Handbook of emotion regulation ( PDFDrive )

Social Baseline Theory 231 References theory: The role of social proximity in emo- tion and economy of action. Social and Per- Ainsworth, M. D. S., Blehar, M. C., Waters, E., sonality Psychology Compass, 5, 976–988. & Wall, S. (1978). Patterns of attachment: A Beckes, L., Coan, J. A., & Hasselmo, K. (2012). psychological study of the strange situation. Familiarity promotes the blurring of self and Hillsdale, NJ: Erlbaum. other in the neural representation of threat. Social Cognitive and Affective Neuroscience. Alexander, R. D. (1974). The evolution of social [E-publication ahead of print] behavior. Annual Review of Ecology and Sys- Berscheid, E. (2003). The human’s greatest tematics, 5, 325–383. strength: Other humans. In U. M. Staudinger (Ed.), A psychology of human strengths: Fun- Allen, J. P., Moore, C., Kuperminc, G., & Bell, K. (1998). Attachment and adolescent psycho- damental questions and future directions for a social functioning. Child Development, 69(5), positive psychology (pp. 37–47). Washington, 1406–1419. DC: American Psychological Association. Bertram, B. C. R. (1978). Living in groups: Pred- Arnsten, A. F. T. (2009). Stress signalling path- ators and prey. In J. R. Krebs & N. B. Davies ways that impair prefrontal cortex structure (Eds.), Behavioural ecology: An evolutionary and function. Nature Reviews Neuroscience, approach (pp. 64–96). Oxford, UK: Black- 10(6), 410–422. well. Bishop, S., Duncan, J., Brett, M., & Lawrence, Baron, R. M., & Kenny, D. A. (1986). The A. D. (2004). Prefrontal cortical function and moderator–­mediator variable distinction in anxiety: Controlling attention to threat-r­ elated social psychological research: Conceptual, stimuli. Nature Neuroscience, 7, 184–188. strategic, and statistical considerations. Jour- Bowlby, J. (1982). Attachment and loss: Vol. 1. nal of Personality and Social Psychology, Attachment (2nd ed.). New York: Basic Books. 51(6), 1173–1182. (Original work published 1969) Brewer, M. B., & Caporeal, L. R. (1990). Selfish Baron, R. S., Cutrona, C. E., Hicklin, D., Rus- genes vs. selfish people: Sociobiology as origin sell, D. W., & Lubaroff, D. M. (1990). Social myth. Motivation and Emotion, 14, 237–243. support and immune function among spouses Brooks, J. C. W., Nurmikko, T. J., Bimson, W. of cancer patients. Journal of Personality and E., Singh, K. D., & Roberts, N. (2002). fMRI Social Psychology, 59(2), 344–352. of thermal pain: Effects of stimulus laterality and attention. NeuroImage, 15, 293–301. Barrett, L. F. (2006). Solving the emotion para- Cancedda, L., Putignano, E., Sale, A., Viegi, A., dox: Categorization and the experience of Berardi, N., & Maffei, L. (2004). Acceleration emotion. Personality and Social Psychology of visual system development by environmen- Review, 10, 20–46. tal enrichment. Journal of Neuroscience, 24, 4840 – 4848. Barrett, L. F., & Bar, M. (2009). See it with Cassidy, J., & Shaver, P. R. (2008). Handbook feeling: Affective predictions during object perception. Philosophical Transactions of the of attachment: Theory, research, and clinical Royal Society B: Biological Sciences, 364, applications (2nd ed.). New York: Guilford 1325–1334. Press. Chugani, H. T. (1998). A critical period of brain Bartels, A., & Zeki, S. (2000). The neural basis development: Studies of cerebral glucose uti- of romantic love. NeuroReport, 11(17), 3829– lization with PET. Preventive Medicine, 27, 3834. 184–188. Clore, G. L., & Tamir, M. (2002). Affect as Bartels, A., & Zeki, S. (2004). The neural corre- embodied information. Psychological Inquiry, lates of maternal and romantic love. NeuroIm- 13, 37–45. age, 21(3), 1155–1166. Coan, J. A. (2008). Toward a neuroscience of attachment. In J. Cassidy & P. R. Shaver Baumeister, R. F., Vohs, K. D., & Tice, D. M. (Eds.), Handbook of attachment: Theory, (2007). The strength model of self-­control. research, and clinical applications (2nd ed., Current Directions in Psychological Science, pp. 241–265). New York: Guilford Press. 16(6), 351–355. Bechara, A. (2011). The somatic marker hypoth- esis and its neural basis: Using past experiences to forecast the future in decision making. In M. Bar (Ed.), Predictions in the brain: Using our past to generate a future (pp. 122–133). New York: Oxford University Press. Beckes, L., & Coan, J. A. (2011). Social baseline

232 SOCIAL ASPECTS Coan, J. A., Beckes, L., & Allen, J. P. (2013). Eisenberger, N. I., Master, S. L., Inagaki, T. K., Childhood maternal support and social capital Taylor, S. E., Shirinyan, D., Lieberman, M. moderate the regulatory impact of social rela- D., et al. (2011). Attachment figures activate a tionships in adulthood. International Journal safety signal-r­elated neural region and reduce of Psychophysiology, 88(3), 224–231. pain experience. Proceedings of the National Academy of Sciences, 108, 11721–11726. Coan, J. A., Schaefer, H. S., & Davidson, R. J. (2006). Lending a hand: Social regulation of Eisenberger, N. I., Taylor, S. E., Gable, S. L., the neural response to threat. Psychological Hilmert, C. J., & Lieberman, M. D. (2007). Science, 17, 1032–1039. Neural pathways link social support to attenu- ated neuroendocrine stress responses. Neuro- Cohen, S., & Hoberman, H. M. (1983). Positive Image, 35(4), 1601–1612. events and social supports as buffers of life change stress. Journal of Applied Social Psy- Elgar, M. A. (1989). Predator vigilance and chology, 13(2), 99–125. group-size in mammals and birds—A critical review of the empirical evidence. Biological Cohen, S., & McKay, G. (1984). Social support, Reviews of the Cambridge Philosophical Soci- stress, and the buffering hypothesis: A theo- ety, 64(1), 13–33. retical analysis. Handbook of Psychology and Health, 4, 253–267. Fitzsimons, G. M., & Finkel, E. J. (2011). Out- sourcing self-r­ egulation. Psychological Sci- Conner, O. L., Siegle, G. J., McFarland, A. M., ence, 22(3), 369–375. Silk, J. S., Ladouceur, C. D., Dahl, R. E., et al. (2012). Mom—It helps when you’re right Fraley, R. C., & Shaver, P. R. (2000). Adult here!: Attenuation of neural stress markers in romantic attachment: Theoretical develop- anxious youths whose caregivers are present ments, emerging controversies, and unan- during fMRI. PloS ONE, 7(12), e50680. swered questions. Review of General Psychol- ogy, 4, 132–154. Davis, L. S. (1984). Alarm calling in Richard- son’s ground squirrels (Spermophilus richard- Franceschini, M. A., Thaker, S., Themelis, G., sonii). Zeitschrift für Tierpsychologie, 66(2), Krishnamoorthy, K. K., Bortfeld, H., Dia- 152–164. mond, S. G., et al. (2007). Assessment of infant brain development with frequency-­ Dedovic, K., Renwick, R., Mahani, N. K., domain near-­infrared spectroscopy. Pediatric Engert, V., Lupien, S. J., & Pruessner, J. C. Research, 61, 546–551. (2005). The Montreal Imaging Stress Task: Using functional imaging to investigate the Francis, S., Rolls, E. T., Bowtell, R., McGlone, effects of perceiving and processing psychoso- F., O’Doherty, J., Browning, A., et al. (1999). cial stress in the human brain. Journal of Psy- The representation of pleasant touch in the chiatry and Neuroscience, 30, 319–325. brain and its relationship with taste and olfac- tory areas. NeuroReport, 10, 453–459. Del Giorno, J. M., Hall, E. E., O’Leary, K. C., Bixby, W. R., & Miller, P. C. (2010). Cogni- Friston, K. (2010). The free-e­nergy principle: A tive function during acute exercise: A test of unified brain theory? Nature Reviews Neuro- the transient hypofrontality theory. Journal of science, 11(2), 127–138. Sport and Exercise Psychology, 32(3), 312– 323. Fuster, J. M. (2002). Frontal lobe and cogni- tive development. Journal of Neurocytology, Dietrich, A. (2009). The transient hypofrontal- 31(3), 373–385. ity theory and its implications for emotion and cognition. In T. McMorris, P. D. Tompo- Gable, S. L., Gonzaga, G. C., & Strachman, A. rowski, & M. Audiffren (Eds.), Exercise and (2006). Will you be there for me when things cognitive function (pp. 69–90). Hoboken, NJ: go right?: Supportive responses to positive Wiley-­Blackwell. event disclosures. Journal of Personality and Social Psychology, 91(5), 904–917. Ehrenberg, M. F., Gearing-S­mall, M., Hunter, M. A., & Small, B. J. (2001). Childcare task Gable, S. L., Reis, H. T., Impett, E. A., & Asher, division and shared parenting attitudes in E. R. (2004). What do you do when things go dual-­earner families with young children. right?: The intrapersonal and interpersonal Family Relations, 50(2), 143–153. benefits of sharing positive events. Journal of Personality and Social Psychology, 87(2), Ehrlich, P. R., & Ehrlich, A. H. (2008). The 228–245. dominant animal: Human evolution and the environment. Washington, DC: Island Press. Gazzola, V., Spezio, M. L., Etzel, J. A., Castelli, F., Adolphs, R., & Keysers, C. (2012). Primary

Social Baseline Theory 233 somatosensory cortex discriminates affective Kable, J. W., & Glimcher, P. W. (2007). The neu- significance in social touch. Proceedings of the ral correlates of subjective value during inter- National Academy of Sciences, 109, E1657– temporal choice. Nature Neuroscience, 10, E1666. 1625–1633. Gillath, O., Bunge, S. A., Shaver, P. R., Wen- delken, C., & Mikulincer, M. (2005). Karau, S. J., & Williams, K. D. (1993). Social Attachment-s­tyle differences in the ability to loafing: A meta-a­ nalytic review and theoreti- suppress negative thoughts: Exploring the neu- cal integration. Journal of Personality and ral correlates. NeuroImage, 28, 835–847. Social Psychology, 65(4), 681–706. Goldstein, R. Z., & Volkow, N. D. (2011). Dys- function of the prefrontal cortex in addiction: Karremans, J. C., Heslenfeld, D. J., van Dillen, L. Neuroimaging findings and clinical implica- F., & Van Lange, P. A. (2011). Secure attach- tions. Nature Reviews Neuroscience, 12(11), ment partners attenuate neural responses 652 – 669. to social exclusion: An fMRI investigation. Halford, G. S., Wilson, W. H., & Phillips, S. (1997). Abstraction: Nature, costs, and ben- International Journal of Psychophysiology, efits. International Journal of Educational 81(1), 44–50. Research, 27(1), 21–35. Kim, S. H., & Hamann, S. (2007). Neural cor- Halford, G. S., Wilson, W. H., & Phillips, S. relates of positive and negative emotion regu- (1998). Processing capacity defined by rela- lation. Journal of Cognitive Neuroscience, tional complexity: Implications for compara- 19(5), 776–798. tive, developmental, and cognitive psychol- Knill, D. C., & Pouget, A. (2004). The Bayesian ogy. Behavioral and Brain Sciences, 21(6), brain: The role of uncertainty in neural coding 803–831. and computation. Trends in Neurosciences, Harlow, H. F. (1958). The nature of love. Ameri- 27(12), 712–719. can Psychologist, 13, 673–685. Kraus, M. W., Huang, C., & Keltner, D. (2010). Hebb, D. O. (1949). The organization of behav- Tactile communication, cooperation, and per- ior. New York: Wiley. formance: An ethological study of the NBA. Heinrichs, M., Baumgartner, T., Kirschbaum, Emotion, 10, 745–749. C., & Ehlert, U. (2003). Social support and Krause, J., & Ruxton, G. (2002). Living in oxytocin interact to suppress cortisol and sub- groups. New York: Oxford University Press. jective responses to psychosocial stress. Bio- Krebs, J. R., & Davies, N. B. (1993). An intro- logical Psychiatry, 54(12), 1389–1398. duction to behavioural ecology (3rd ed.). Mal- Higashi, M., & Yamamura, N. (1993). What den, MA: Blackwell. determines animal group size?: Insider–­ Kurzban, R. (2010). Does the brain consume outsider conflict and its resolution. American additional glucose during self-­control tasks? Naturalist, 142(3), 553–563. Evolutionary Psychology, 8(2), 244–259. Hill, K. R., Walker, R. S., Bozicevic, M., Eder, J., Langer, S. K. (1974). Mind: An essay on human Headland, T., Hewlett, B., et al. (2011). Co-­ feeling (Vol. 2). Baltimore: Johns Hopkins residence patterns in hunter–­gatherer societies University Press. show unique human social structure. Science, Langston, C. A. (1994). Capitalizing on and cop- 331(6022), 1286–1289. ing with daily-life events: Expressive responses Hofer, M. A. (2006). Psychobiological roots of to positive events. Journal of Personality and early attachment. Current Directions in Psy- Social Psychology, 67(6), 1112–1125. chological Science, 15, 84–88. Larose, S., Bernier, A., Soucy, N., & Duchesne, Holt-L­ unstad, J., Smith, T. B., & Layton, J. B. S. (1999). Attachment style dimensions, net- (2010). Social relationships and mortality risk: work orientation and the process of seeking A meta-­analytic review. PLoS Medicine, 7, help from college teachers. Journal of Social e1000316. and Personal Relationships, 16, 225–247. Hughes, A. E., Crowell, S. E., Uyeji, L., & Coan, Lieberman, M. D. (2007). Social cognitive neu- J. A. (2012). A developmental neuroscience of roscience: A review of core processes. Annual borderline pathology: Emotion dysregulation Review of Psychology, 58, 259–289. and social baseline theory. Journal of Abnor- Lindquist, K. A., & Barrett, L. F. (2008). Con- mal Child Psychology, 40, 21–33. structing emotion. Psychological Science, 19, 898–903. Lipina, S. J., & Colombo, J. A. (2009). Poverty and brain development during childhood: An

234 SOCIAL ASPECTS approach from cognitive psychology and neu- (2007). When fear is near: Threat imminence roscience. Washington, DC: American Psy- elicits prefrontal–­periaqueductal gray shifts in chological Association. humans. Science, 317(5841), 1079–1083. MacArthur, R. H., & Pianka, E. R. (1966). On Moll, H., & Tomasello, M. (2007). Cooperation optimal use of a patchy environment. Ameri- and human cognition: The Vygotskian intelli- can Naturalist, 100, 603–609. gence hypothesis. Philosophical Transactions Main, M. (1996). Introduction to the special sec- of the Royal Society B: Biological Sciences, tion on attachment and psychopathology: 2. 362(1480), 639–648. Overview of the field of attachment. Journal Monson, C. M., Fredman, S. J., Macdonald, A., of Consulting and Clinical Psychology, 64, Pukay-M­ artin, N. D., Resick, P. A., & Schnurr, 237–243. P. P. (2012). Effect of cognitive-b­ehavioral McComb, K., Moss, C., Durant, S. M., Baker, L., couple therapy for PTSD: A randomized con- & Sayialel, S. (2001). Matriarchs as reposito- trolled trial. Journal of the American Medical ries of social knowledge in African elephants. Association, 308, 700–709. Science, 292, 491–494. Moore, L. J., Vine, S. J., Wilson, M. R., & Free- McGowan, P. O., Sasaki, A., D’Alessio, A. C., man, P. (2012). The effect of challenge and Dymov, S., Labonté, B., Szyf, M., et al. (2009). threat states on performance: An examination Epigenetic regulation of the glucocorticoid of potential mechanisms. Psychophysiology, receptor in human brain associates with child- 49(10), 1417–1425. hood abuse. Nature Neuroscience, 12, 342– Morrison, I., Löken, L. S., & Olausson, H. 348. (2010). The skin as a social organ. Experimen- McNally, L., Brown, S. P., & Jackson, A. L. tal Brain Research, 204, 305–314. (2012). Cooperation and the evolution of intel- Muraven, M., & Baumeister, R. F. (2000). Self-­ ligence. Proceedings of the Royal Academy B: regulation and depletion of limited resources: Biological Sciences, 279(1740), 3027–3034. Does self-c­ontrol resemble a muscle? Psycho- Michelena, P., Pillot, M. H., Henrion, C., Toulet, logical Bulletin, 126(2), 247–259. S., Boissy, A., & Bon, R. (2012). Group size Naaman, S., Radwan, K., & Johnson, S. (2009). elicits specific physiological response in herbi- Coping with early breast cancer: Couple vores. Biology Letters, 8(4), 537–539. adjustment processes and couple-b­ ased inter- Mikulincer, M., & Florian, V. (1998). The rela- vention. Psychiatry: Interpersonal and Bio- tionship between adult attachment styles and logical Processes, 72(4), 321–345. emotional and cognitive reactions to stressful Paxton, J. L., Barch, D. M., Racine, C. A., & events. In J. A. Simpson & W. S. Rholes (Eds.), Braver, T. S. (2008). Cognitive control, goal Attachment theory and close relationships maintenance, and prefrontal function in (pp. 143–165). New York: Guilford Press. healthy aging. Cerebral Cortex, 18(5), 1010– Mikulincer, M., & Orbach, I. (1995). Attach- 1028. ment styles and repressive defensiveness: The Posner, M. I., & Rothbart, M. K. (2007). accessibility and architecture of affective Research on attention networks as a model memories. Journal of Personality and Social for the integration of psychological science. Psychology, 68, 917–925. Annual Review of Psychology, 58, 1–23. Mikulincer, M., & Shaver, P. R. (2005). Attach- Prinstein, M. J., & La Greca, A. M. (2004). ment theory and emotions in close relation- Childhood peer rejection and aggression as ships: Exploring the attachment-r­elated predictors of adolescent girls’ externalizing dynamics of emotional reactions to relational and health risk behaviors: A 6-year longitudi- events. Personal Relationships, 12, 149–168. nal Study. Journal of Consulting and Clinical Mikulincer, M., & Shaver, P. R. (2007). Attach- Psychology, 72, 103–112. ment in adulthood: Structure, dynamics, and Pulliam, H. R., Pyke, G. H., & Caraco, T. (1982). change. New York: Guilford Press. The scanning behavior of juncos: A game–­ Mikulincer, M., Shaver, P. R., & Pereg, D. theoretical approach. Journal of Theoretical (2003). Attachment theory and affect regula- Biology, 95(1), 89–103. tion: The dynamics, development, and cogni- Reichardt, L. F. (2006). Neurotrophin-r­egulated tive consequences of attachment-r­elated strat- signalling pathways. Philosophical Transac- egies. Motivation and Emotion, 27, 77–102. tions of the Royal Society B: Biological Sci- Mobbs, D., Petrovic, P., Marchant, J. L., Has- ences, 361, 1545–1564. sabis, D., Weiskopf, N., Seymour, B., et al. Reis, H. T., Smith, S. M., Carmichael, C. L.,

Social Baseline Theory 235 Caprariello, P. A., Tsai, F. F., Rodrigues, A., self-­control effort. International Journal of et al. (2010). Are you happy for me?: How Psychophysiology, 71(2), 91–96. sharing positive events with others provides Sherwood, C. C., Stimpson, C. D., Raghanti, M. personal and interpersonal benefits. Journal A., Wildman, D. E., Uddin, M., Grossman, L. of Personality and Social Psychology, 99(2), I., et al. (2006). Evolution of increased glia–­ 311–329. neuron ratios in the human frontal cortex. Rholes, W. S., Simpson, J. A., & Grich Stevens, J. (1998). Attachment orientations, social sup- Proceedings of the National Academy of Sci- port, and conflict resolution in close relation- ences, 103(37), 13606–13611. ships. In J. A. Simpson & W. S. Rholes (Eds.), Siegfried, W. R., & Underhill, L. G. (1975). Flocking as an anti-­predator strategy in doves. Attachment theory and close relationships Animal Behaviour, 23, 504–508. (pp. 166–188). New York: Guilford Press. Smith, E. A. (2003). Human cooperation: Per- Richerson, P. J., Bettinger, R. L., & Boyd, R. spectives from behavioral ecology. In P. Ham- (2005). Evolution on a restless planet: Were merstein (Ed.), Genetic and cultural evolution environmental variability and environmental of cooperation (pp. 401–427). Cambridge, change major drivers of human evolution? In MA: MIT Press. F. M. Wuketits & C. Antweiler (Eds.), Hand- Sokoloff, L., Mangold, R., Wechsler, R. L., Ken- nedy, C., & Kety, S. S. (1955). The effect of book of evolution: The evolution of living sys- mental arithmetic on cerebral circulation and tems (Vol. 2, pp. 223–242). New York: Wiley. metabolism. Journal of Clinical Investigation, Roberts, G. (1996). Why individual vigilance 34(7, Pt. 1), 1101–1108. declines as group size increases. Animal Stark, R., Schienle, A., Walter, B., Kirsch, P., Behaviour, 51, 1077–1086. Sammer, G., Ott, U., et al. (2003). Hemody- Rogers, S. J., & DeBoer, D. D. (2001). Changes namic responses to fear and disgust-­inducing in wives’ income: Effects on marital happi- pictures: An fMRI study. International Jour- ness, psychological well-being, and the risk nal of Psychophysiology, 50, 225–234. of divorce. Journal of Marriage and Family, Stefanucci, J. K., Gagnon, K. T., & Lessard, D. 63(2), 458–472. A. (2011). Follow your heart: Emotion adap- Salinas, E. (2011). Prior and prejudice. Nature tively influences perception. Social and Per- Neuroscience, 14, 943–945. sonality Psychology Compass, 5, 296–308. Schilbach, L., Timmermans, B., Reddy, V., Cos- Stefanucci, J. K., Proffitt, D. R., Banton, T., & tall, A., Bente, G., Schlicht, T., et al. (in press). Epstein, W. (2005). Distances appear different Toward a second-p­ erson neuroscience. Behav- on hills. Attention, Perception, and Psycho- ioral and Brain Sciences. physics, 67, 1052–1060. Schmidhempel, P., Kacelnik, A., & Houston, A. Stefanucci, J. K., Proffitt, D. R., Clore, G. L., I. (1985). Honeybees maximize efficiency by & Parekh, N. (2008). Skating down a steeper not filling their crop. Behavioral Ecology and slope: Fear influences the perception of geo- Sociobiology, 17(1), 61–66. graphical slant. Perception, 37, 321–323. Schnall, S., Harber, K. D., Stefanucci, J. K., & Tomaka, J., Blascovich, J., Kelsey, R. M., & Leit- Proffitt, D. R. (2008). Social support and ten, C. L. (1993). Subjective, physiological, the perception of geographical slant. Journal and behavioral effects of threat and challenge of Experimental Social Psychology, 44(5), appraisal. Journal of Personality and Social 1246–1255. Psychology, 65, 248–260. Schwarz, N., & Clore, G. L. (1983). Mood, Tomasello, M., Carpenter, M., Call, J., Behne, misattribution, and judgments of well-being: T., & Moll, H. (2005). Understanding and Informative and directive functions of affec- sharing intentions: The origins of cultural tive states. Journal of Personality and Social cognition. Behavioral and Brain Sciences, 28, Psychology, 45, 513–523. 675 – 690. Shallcross, S. L., Howland, M., Bemis, J., Simp- Townsend, A. L., & Franks, M. M. (1995). Bind- son, J. A., & Frazier, P. (2011). Not “capitaliz- ing ties: Closeness and conflict in adult chil- ing” on social capitalization interactions: The dren’s caregiving relationships. Psychology role of attachment insecurity. Journal of Fam- and Aging, 10(3), 343–351. ily Psychology, 25(1), 77–85. Uchino, B. N., Cacioppo, J. T., & Kiecolt-G­ laser, Sheppes, G., Catran, E., & Meiran, N. (2009). J. K. (1996). The relationship between social Reappraisal (but not distraction) is going to support and physiological processes: A review make you sweat: Physiological evidence for

236 SOCIAL ASPECTS with emphasis on underlying mechanisms and D’Alessio, A. C., Sharma, S., Seckl, J. R., et al. implications for health. Psychological Bulle- (2004). Epigenetic programming by maternal tin, 119, 488–531. behavior. Nature Neuroscience, 7, 847–854. van IJzendoorn, M. H., Bakermans-K­ ranenburg, Woolley, A. W., Chabris, C. F., Pentland, A., M. J., & Ebstein, R. P. (2011). Methylation Hashmi, N., & Malone, T. W. (2010). Evi- matters in child development: Toward devel- dence for a collective intelligence factor in the opmental behavioral epigenetics. Child Devel- performance of human groups. Science, 330, opment Perspectives, 5, 305–310. 686 – 688. Vohs, K. D., & Heatherton, T. F. (2000). Self-­ Younger, J., Aron, A., Parke, S., Chatterjee, N., regulatory failure: A resource-­depletion & Mackey, S. (2010). Viewing pictures of a approach. Psychological Science, 11(3), 249– romantic partner reduces experimental pain: 254. Involvement of neural reward systems. PloS Weaver, I. C. G., Cervoni, N., Champagne, F. A., ONE, 5, e13309.

Chapter 15 Adult Attachment and Emotion Regulation Phillip R. Shaver Mario Mikulincer In the past 30 years, attachment theory ideas (as elaborated and tested initially by (Bowlby, 1973, 1980, 1969/1982) has Ainsworth, Blehar, Waters, & Wall, 1978) become one of the most influential concep- have been expanded, tested, and organized tual frameworks for understanding emotion by social–­personality psychologists into a regulation. Although Bowlby did not devote theoretical model of the attachment system much attention to abstract theorizing about in adulthood (e.g., Mikulincer & Shaver, emotion itself (he included only a single brief 2007a; Shaver & Mikulincer, 2002). The chapter about it in Volume 1 of his Attach- model focuses on the emotion-­regulatory ment and Loss trilogy), his writings were function of the attachment system and motivated by clinical and ethological obser- explains many of the emotional corre- vations of humans and other primates who lates and consequences of normative and were experiencing, expressing, and regu- individual-­differences aspects of attach- lating emotions such as affection, anxiety, ment system functioning. In this chapter, we anger, grief, and despair. He was especially elaborate on these normative and individual-­ interested in the anxiety-­buffering func- differences aspects, and review studies that tion of close relationships and the capacity have examined the links between attachment for dysfunctional relationships to generate system activation and emotion regulation, negative emotions, and, in the extreme, to and the ways in which individual differences precipitate debilitating forms of psychopa- in attachment are reflected in patterns of thology. Bowlby (1973, 1980) described and coping with attachment-i­rrelevant stressful conceptualized the relatively stable individ- events and attachment-­related stressors (e.g., ual differences in emotion regulation that a partner’s hurtful behavior, the breakup of emerge from prolonged reliance on particu- a relationship, the death of a relationship lar attachment figures, people who provide partner). either adequate or inadequate protection, safety, support, and guidance concerning Attachment Theory: Basic Concepts emotions and emotion regulation. Bowlby (1969/1982) claimed that human beings are born with an innate psychobio- With the accumulation of empirical logical system (the attachment behavioral knowledge about what Bowlby called the system) that motivates them to seek proxim- “attachment behavioral system” and about individual differences in attachment orienta- tions in childhood and adulthood, Bowlby’s 237

238 SOCIAL ASPECTS ity to significant others (attachment figures) 2002). In contrast, deactivation involves the in times of need. This system accomplishes inhibition of proximity-s­eeking inclinations basic regulatory functions (protection from and actions, suppression or discounting of threats and alleviation of distress) in human threats that might activate the attachment beings of all ages, but it is most directly system, down-­regulation of both negative observable during infancy and early child- and positive affective states, and determi- hood (Bowlby, 1988). According to Bowlby nation to handle stressors alone (a defen- (1969/1982), attachment figures can pro- sive stance that Bowlby [1969/1982] called vide a physical and emotional safe haven “compulsive self-­reliance”). People who rely that facilitates the down-­regulation of nega- on these strategies tend to maximize auton- tive affective states, and a secure base that omy and distance from relationship partners encourages positive affective states condu- and experience discomfort with closeness cive to exploration and learning, thus sup- and intimacy. porting the development of knowledge and skills. When testing this theory in studies of adults, most researchers have focused on the Bowlby (1973) also described important organized pattern of relational expectations, individual differences in attachment system emotions, and behavior that results from a functioning shaped by the responsiveness history of interactions with attachment and supportiveness of attachment figures. figures—­often called “attachment style” by Interactions with attachment figures who social psychologists (e.g., Fraley & Shaver, are responsive in times of need facilitate 2000). Psychometric research (e.g., Bren- the optimal functioning of the attachment nan, Clark, & Shaver, 1998) has shown that system and promote a sense of attachment attachment styles can be measured in terms security. This deep and pervasive sense of of two independent dimensions: attachment security is based on implicit beliefs that anxiety and attachment-r­elated avoidance. the world is generally safe, that attachment A person’s position on the attachment anxi- figures are helpful when called upon, and ety dimension indicates the degree to which that it is possible to explore the environ- he or she worries that a partner will not be ment curiously and to engage enjoyably with responsive in times of need. A person’s posi- other people without undue fear. This sense tion on the avoidance dimension indicates of security is rooted in positive mental repre- the extent to which he or she distrusts rela- sentations of self and others, which Bowlby tionship partners’ goodwill and their will- called internal working models. ingness and capacity to help without causing further distress. People who score low on Unfortunately, when attachment figures both dimensions are said to be secure with are not reliably available and supportive, respect to attachment. A person’s location and when they fail to provide adequate in the continuous two-d­ imensional anxiety-­ relief from distress, children who are depen- by-a­voidance space can be measured with dent on them may form negative work- reliable and valid self-r­eport scales (e.g., ing models of self and others and develop Brennan et al., 1998) and is associated in defensive secondary attachment strategies theoretically predictable ways with many that involve hyperactivation or deactiva- aspects of psychological adjustment (see tion of the attachment system (e.g., Cassidy Mikulincer & Shaver, 2007a, for a review). & Kobak, 1988). Hyperactivation (which Bowlby, 1982, called “protest”) is charac- Attachment styles begin to develop in terized by intense efforts to attain proximity interactions with primary caregivers during to attachment figures and insistent attempts early childhood, as a large body of research to induce a relationship partner, viewed as demonstrates (see Cassidy & Shaver, 2008, insufficiently available or responsive, to for an anthology of reviews), but Bowlby provide more satisfying and reassuring care (1988) claimed that relational experiences and support. Hyperactivating strategies throughout life can move a person from include clinging, controlling, and coercive one region to another in the anxiety-­by-­ behaviors; cognitive and behavioral efforts avoidance conceptual space. Moreover, to establish physical contact and a sense although attachment style is often mea- of “oneness”; and up-r­egulation of nega- sured as a single global orientation toward tive affective states (Shaver & Mikulincer, close relationships, a person’s orientation to

Attachment and Emotion Regulation 239 attachment is rooted in a complex cognitive phrased by us as follows: “If I encounter and affective associative neural network that an obstacle and/or become distressed, then includes both secure and insecure episodic I can approach a significant other for help. and semantic memories and mental repre- He or she is likely to be available and sup- sentations (Mikulincer & Shaver, 2007a). portive. I will experience relief and comfort Indeed, a growing body of research shows as a result of proximity to this person, and that attachment style can change, subtly I can then return to other activities.” Once or dramatically, depending on current con- activated, this script can, by itself, mitigate text and recent relational experiences (see distress, promote optimism and hope, and Mikulincer & Shaver, 2007b, for a review). help a person cope effectively with life’s For example, we have found that priming inevitable difficulties (Mikulincer, Shaver, thoughts of an available and supportive Sapir-Lavid, & Avihou-Kanza, 2009). attachment figure leads people who score relatively high on scales tapping attach- Adult attachment researchers have ment insecurities momentarily to behave designed experimental procedures to exam- like more secure people (e.g., Mikulincer & ine these normative regulatory properties Shaver, 2001). of the attachment system. For example, in a series of experiments, Mikulincer, Gil- In the next two sections, we present both lath, and Shaver (2002) showed that men- theoretical ideas and empirical evidence tal representations of attachment figures concerning the link between attachment (e.g., names of security-e­nhancing attach- and emotion regulation. In the first section, ment figures) are automatically activated in we focus on normative aspects of this link a person’s mind when he or she is exposed and the emotion regulation functions of to threatening stimuli, even if that exposure the attachment behavioral system. That is, is unconscious (i.e., primed subliminally). we discuss and review empirical evidence Specifically, when a threat-­related word concerning the normative activation of the (e.g., death) was presented subliminally on attachment system in response to threats a computer screen, participants were faster and the down-­regulation of distress caused to detect the name of one of their attachment by optimal functioning of the system. In the figures when it appeared on the screen and second section, we focus on individual dif- slower to name the color in which such names ferences and review empirical evidence con- were printed on the screen—­an indication cerning the up-r­egulation of negative affect that the names had been automatically and caused by anxious attachment, and the unconsciously activated in memory (Miku- ways in which avoidant attachment leads to lincer et al., 2002). In other words, threats, down-­regulation of both negative and posi- even when arising unconsciously, can auto- tive affective states. matically activate mental representations of security providers. Similar findings have Normative Aspects of the Link been obtained in experiments examining between Attachment automatic activation of representations of and Emotion Regulation symbolic sources of security, such as a per- According to attachment theory (Bowlby, son’s pet or God, following a threat (Gran- 1969/1982), the attachment system evolved qvist, Mikulincer, Gurwitz, & Shaver, 2012; because it increased the likelihood that off- Zilcha-Mano, Mikulincer, & Shaver, 2011). spring would survive until they were old enough to reproduce and also taught chil- In another set of studies of what we call dren how to regulate emotions effectively. security priming, Mikulincer, Hirschberger, This system is automatically activated by Nachmias, and Gillath (2001) showed that external threats or internal sources of dis- activation of representations of security-­ tress and, when it functions appropriately, enhancing attachment figures can automati- leads to emotional security. Its optimal cally infuse a previously neutral stimulus functioning is associated with a relational with positive affect. For example, subliminal if–then script, which Waters and Waters presentation of the names of people who were (2006) called a secure-b­ase script, para- nominated by study participants as attach- ment figures, compared with others who were not nominated as attachment figures, led to greater liking of previously unfamil-

240 SOCIAL ASPECTS iar stimuli. Moreover, subliminal exposure ing and help a person maintain a sense of to names of attachment figures eliminated well-being even in stressful conditions or the detrimental effects that threats other- difficult periods of life (Shaver & Miku- wise had on liking for previously neutral lincer, 2002). Moreover, these attachment stimuli. These effects of security priming on strategies encourage repeated activation positive affect have been replicated in sub- and suppression of negative emotions and sequent studies (see Mikulincer & Shaver, continued reliance on negative or distorted 2007b, for a review). In addition, Eisen- mental representations of self and others, berger and colleagues (2011; Master et al., which can erode and destroy mental health 2009) found that viewing a photograph of (Mikulincer & Shaver, 2007a). one’s romantic partner (vs. a stranger or an object) reduced one’s subjective experience The early attachment experiences of inse- of pain while receiving thermal stimulation cure people (whether anxious, avoidant, at levels slightly higher than the pain thresh- or both) often involve unstable and inad- old. Moreover, Selcuk, Zayas, Günaydin, equate distress regulation (Bowlby, 1973), Hazan, and Kross (2012) found that both which interferes with the development of explicit and implicit priming of attachment inner resources needed for coping success- figure representations speeded up emotional fully with stressors. This impairment is recovery and reduced negative thoughts particularly likely during prolonged, highly after recalling an upsetting experience. demanding stressful experiences that require active support seeking and actual confronta- Considering all such findings, we con- tion with a problem (Berant, Mikulincer, & clude that people automatically search Shaver, 2008). In such cases, anxious hyper- for internal representations of security-­ activating strategies can become extreme, enhancing attachment figures during times damaging not only a person’s own mental of stress, and mental activation of these health but that of key relationship partners. representations produces positive emotions Moreover, avoidant defenses can collapse, (e.g., relief, satisfaction, gratitude, love) that resulting in a marked decline in psychologi- facilitate effective coping and restore emo- cal well-being. tional equanimity. That is, actual or sym- bolic interactions with available and sup- When regulating their emotions, avoid- portive attachment figures, and the resulting ant people attempt to block or inhibit any sense of safety, can be viewed as psycho- emotional state that is incongruent with logical resources for dealing with problems the goal of keeping their attachment system and adversities, and sustaining well-being deactivated (Mikulincer & Shaver, 2007a). and mental health (Mikulincer & Shaver, These inhibitory efforts are directed mainly 2007a). at fear, anxiety, anger, sadness, shame, guilt, and distress, because these emotions are Attachment‑Related Individual associated with threats and feelings of vul- Differences in Emotion Regulation nerability. In addition, anger often implies Theoretical Account emotional involvement or investment in a According to Bowlby (1973), disruptions in relationship, and such involvement is incon- the sense of attachment security are viewed gruent with avoidant people’s preference as risk factors for emotional problems and for independence and self-r­eliance (Cassidy, psychopathology. Although secondary 1994). Avoidant individuals also attempt attachment strategies (anxious hyperacti- to block or inhibit emotional reactions to vation and avoidant deactivation) are ini- potential or actual threats to attachment tially adaptive, in the sense that they adjust figure availability (rejection, betrayal, sepa- a child’s behavior to the requirements of ration, loss), because such threats are direct an inconsistently available or consistently triggers of attachment system activation. distant or unavailable attachment figure, Like secure people, avoidant ones attempt to they are maladaptive when used in later down-­regulate threat-­related emotions. But relationships in which support seeking and whereas secure people’s regulatory attempts relational interdependence could be reward- usually promote communication, compro- mise, and relationship maintenance, avoid- ant people’s efforts are aimed mainly at keeping the attachment system deactivated,

Attachment and Emotion Regulation 241 regardless of the deleterious effects this can & Shaver, 2007a). Another hyperactivating have on a relationship. strategy is to intensify negative emotions by favoring an approach, counterphobic ori- Deactivating strategies cause people to entation toward threatening situations or avoid noticing their own emotional reac- making self-d­ efeating decisions and taking tions. Avoidant individuals often deny or ineffective actions that are likely to end in suppress emotion-r­elated thoughts and failure. All of these strategies create a self-­ memories, divert attention from emotion-­ amplifying cycle of distress even after a related material, suppress emotion-r­elated threat objectively recedes. action tendencies, or inhibit or mask ver- bal and nonverbal expressions of emotion Empirical Evidence (Mikulincer & Shaver, 2007a). By averting There is now a large body of evidence sup- the conscious experience and expression of porting the hypothesized links between unpleasant emotions, avoidant individuals attachment-­related avoidance and emotional make it less likely that emotional experi- inhibition, and between attachment anxiety ences will be integrated into their cognitive and distress intensification. In particular, structures and that such feelings and mental theory-c­ongruent findings have been found structures will be used effectively in informa- in studies of responses to both attachment-­ tion processing and social behavior. Bowlby irrelevant stressful events and attachment-­ (1980) described this strategy as “defensive related stressors (e.g., a partner’s hurtful exclusion” and the creation of “segregated behavior, a relationship breakup, or the mental systems.” death of a relationship partner). Unlike secure and avoidant people, who Attachment Orientations and Responses tend to view negative emotions as goal-­ to Stressful Events incongruent states that should either be Several studies have examined attachment managed effectively or suppressed, anx- style differences in the ways people appraise, iously attached individuals tend to perceive cope with, and emotionally and physiologi- these emotions as congruent with attach- cally react to attachment-­irrelevant stress- ment goals, and they may seek to sustain and ful events (i.e., events that have no direct even exaggerate them. Attachment-a­ nxious implications for one’s close relationships or people are guided by an unfulfilled wish attachment figures). These attachment style to cause attachment figures to pay more differences have been examined in response attention and provide more reliable protec- to a wide variety of stressful experiences, tion and support (Cassidy & Kobak, 1988; such as combat training (e.g., Mikulincer Mikulincer & Shaver, 2007a). Therefore, & Florian, 1995), transition to college (e.g., they tend to exaggerate the presence and Lopez & Gormley, 2002), abortion (e.g., seriousness of threats and to overemphasize Cozzarelli, Sumer, & Major, 1998), infertil- their sense of helplessness and vulnerability, ity (e.g., Amir, Horesh, & Lin-Stein, 1999), because signs of weakness and neediness can pregnancy (e.g., Mikulincer & Florian, sometimes elicit attachment figures’ atten- 1999), workload (e.g., Raskin, Kummel, & tion and care (Cassidy & Berlin, 1994). Bannister, 1998), financial problems (e.g., Bartley, Head, & Stansfeld, 2007), and job How is anxious hyperactivation sustained? loss (e.g., Hobdy et al., 2007). They have One method is to exaggerate the appraisal also been examined in studies that did not process, perceptually heightening the threat- focus on a specific stressful event but asked ening aspects of even fairly benign events, participants to report on any major stress- to hold pessimistic beliefs about one’s ability ors recently experienced (e.g., Holmberg, to manage distress, and to attribute threat-­ Lomore, Takacs, & Price, 2011). In addi- related events to uncontrollable causes or tion, the link between attachment orienta- global personal inadequacies (Mikulincer & tions and stress-­related reactions has been Shaver, 2007a). Another method is to attend examined in response to several laboratory-­ to internal indicators of distress (Cassidy induced stressors, such as aversive noise (e.g., & Kobak, 1988). This includes hypervigi- lant attention to the physiological aspects of emotional states, heightened recall of threat-r­elated experiences, and rumination on real and potential threats (Mikulincer

242 SOCIAL ASPECTS Quirin, Pruessner, & Kuhl, 2008), the Trier Recent studies have found that attachment-­ Social Stress Test (e.g., Smeets, 2010), and related differences in coping are reflected in difficult cognitive tasks (e.g., Kidd, Hamer, physiological responses to stressful events. & Steptoe, 2011). For example, two experiments (Diamond, Hicks, & Otter-­Henderson, 2006; Maunder, With regard to the cognitive appraisal Lancee, Nolan, Hunter, & Tannenbaum, of stressful events, attachment anxiety has 2006) exposed participants to various labo- been associated with distress-i­ntensifying ratory stressors (e.g., recalling a stressful sit- appraisals—i­n which stressful events are uation, performing demanding math tasks) appraised as threats rather than challenges and found that self-­reports of avoidant and one’s coping resources are viewed attachment were associated with heightened as deficient (e.g., Cozzarelli et al., 1998; physiological reactivity: decreased heart rate Mikulincer & Florian, 1998). For avoidant variability, increased skin conductance, and attachment, however, the findings are more heightened blood pressure. With regard to complex. Whereas avoidant attachment is attachment anxiety, Maunder et al. found associated with appraising stressful events that attachment-a­nxious people reported in more threatening terms, it is not associ- higher levels of distress but did not exhibit ated with appraisals of coping resources heightened physiological reactivity, perhaps (e.g., Williams & Riskind, 2004). This dis- suggesting, as mentioned earlier, that anx- sociation might reflect avoidant individuals’ ious people exaggerate their own distress. reluctance to recognize that they are vulner- In addition, four studies have assessed able and weak, and therefore need to rely attachment style differences in the activ- on others for assistance and support (Miku- ity of the hypothalamic–p­ituitary–­adrenal lincer & Shaver, 2007a). (HPA) axis, which can be measured by salivary cortisol levels during and following There is also theoretically consistent evi- a laboratory-­induced stressor. One study dence concerning attachment-­related dif- found that avoidant attachment was associ- ferences in self-­reports of ways of coping ated with increased levels of salivary corti- with stressful events. Whereas attachment sol (Kidd et al., 2011); another study (Qui- anxiety is associated with emotion-­focused rin et al., 2008) found a positive association coping, such as wishful thinking, self-blame, between attachment anxiety and heightened and distress-r­elated rumination, avoidant cortisol reactivity; and two studies found no attachment is associated with reliance on significant association between attachment distancing strategies, such as stress denial, orientations and cortisol levels (Ditzen et diversion of attention, and behavioral or al., 2008; Smeets, 2010). These inconsisten- cognitive disengagement (e.g., Holmberg et cies might have been due, in part, to varia- al., 2011; Zhang & Labouvie-V­ ief, 2004). tions in the laboratory-i­nduced stressors (e.g., aversive noise, the Trier Social Stress Several adult attachment studies have Test) and participants’ age (young adults, involved asking participants to report on midlife adults). More systematic research is their psychological distress or well-being needed to determine the presence or absence during and following stressful events. of links between attachment insecurities Overall, findings indicate that attach- and HPA dysregulation following stressful ment insecurities—a­nxiety, avoidance, or events. both—are associated with self-r­eports of There is also preliminary evidence that heightened distress and deteriorated well- an association between attachment inse- being (see Mikulincer & Shaver, 2007a, for curities and psychological distress follow- a review). Moreover, some of the studies ing stressful events can be observed in compared the emotional reactions of secure brain responses. Using event-­related func- and insecure people undergoing stressful tional magnetic resonance imaging (fMRI), experiences with those of controls, reveal- Lemche et al. (2006) found that self-r­eports ing that stressful events arouse distress of attachment anxiety or avoidance were mainly among insecurely attached people. associated with heightened activation in For secure people, there is often no notable bilateral amygdalae to a stressful stimulus. difference in emotion between neutral and That is, less secure people tended to react to stressful situations (e.g., Amir et al., 1999). That is, secure people seem to be relatively calm under stressful conditions.

Attachment and Emotion Regulation 243 stress with increased amygdala activity—­a partner’s transgressions. Moreover, whereas neural indication of distress-­related arousal. secure individuals tended to respond to rejec- tion threats by adopting approach-o­ riented Considering longer-t­erm neural effects goals that improved relationship quality, of attachment insecurities, Quirin, Gil- less secure people tended to respond to such lath, Pruessner, and Eggert (2010) found threats by decreasing approach motivation that self-­reports of attachment anxiety and and emphasizing avoidance goals (Park, avoidance were associated with reduced 2010). In a test of responses to a hurtful hippocampal cell density, which was asso- interaction, Perunovic and Holmes (2008) ciated with poorer emotion regulation. In found that more secure individuals reported particular, attachment-­related avoidance inhibition of negative impulses and the was associated with bilateral hippocampal adoption of relationship-e­nhancing behav- cell reduction, and attachment anxiety was ior even under time pressure, implying that significantly related to reduced cell concen- for them accommodation was automatic. tration in the left hippocampus. These find- ings are compatible with a neurotoxic model Dewitte and De Houwer (2008; Dewitte, of stress-i­nduced cell reduction in the hip- Koster, De Houwer, & Buysse, 2007) used pocampus, contributing to poorer emotion a classic dot-probe task and found that regulation abilities in individuals with inse- both attachment anxiety and avoidance cure attachment orientations. were associated with attentional avoid- ance of signs of a partner’s hurtful behavior Attachment Orientations and Responses (attachment-t­hreat words and angry faces). to a Partner’s Hurtful Behaviors In addition, using the same dot-probe task, Explicit or implicit signs of a partner’s disap- Dewitte, De Houwer, Koster, and Buysse proval, criticism, rejection, or betrayal can (2007) found that attachment anxiety was evoke hurt feelings, arouse hostility toward associated with hypervigilance regarding a partner, and even destroy a relationship. attachment figures’ names in both threaten- However, the ways people react to these ing and positive relational situations. This relational stressors seem to depend on their finding is conceptually similar to Miku- attachment orientation. There is accumu- lincer et al.’s (2002) finding of faster cogni- lating evidence that insecure people tend to tive processing of attachment figure’s names react more intensely than secure people—­ regardless of context. cognitively, emotionally, and behavior- ally—t­o actual or imagined partner disap- Attachment insecurities also tend to proval, criticism, rejection, or betrayal (e.g., inhibit forgiveness—­one of the most effec- Besser & Priel, 2009, 2010; Carnelley, Israel, tive accommodation responses for improv- & Brennan, 2007). For example, Dewitte, ing relationship quality and reestablishing De Houwer, Goubert, and Buysse (2010) relational harmony following a partner’s found that whereas attachment anxiety was negative behavior (e.g., Burnette, Taylor, related to greater physiological arousal and Worthington, & Forsyth, 2007; Yárnoz- stronger negative affect following relational Yaben, 2009). Mikulincer, Shaver, and stress, attachment-­related avoidance pre- Slav (2006) found that less secure people dicted withdrawal responses. were more inclined to report intense feel- ings of vulnerability or humiliation and a Using Rusbult, Verette, Whitney, Slovik, strong sense of relationship deterioration and Lipkus’s (1991) typology of accom- when “forgiving” a partner. In other words, modation responses to a partner’s negative attachment insecurities were associated behavior, several studies (e.g., Gaines et al., with a more negative conception and experi- 1997; Scharfe & Bartholomew, 1995) found ence of forgiveness. Burnette, Davis, Green, that, compared with insecure people, secure Worthington, and Bradfield (2009) provided ones were more likely to rely on “voice” evidence concerning potential mediators (an active, relational approach to solving a of such effects: Whereas the link between problem) and “loyalty” (understanding the attachment anxiety and forgiveness was temporary nature of a partner’s behavior mediated by excessive rumination on rela- and working with him or her to improve)— tional injuries, the link between attachment-­ the two more accommodative responses to a related avoidance and forgiveness was medi- ated by lack of empathy.

244 SOCIAL ASPECTS Beyond these associations between dispo- tions. These typical correlational findings were dramatically reduced in size (most sitional measures of attachment insecurities approached zero) in the security-p­riming and deficient forms of forgiveness, there is condition. In other words, security priming increasing evidence that state-like senses of reduced the tendency of avoidant people to security or insecurity can alter the tendency rely on cool hostility or denial, and the ten- to forgive a hurtful partner. For example, dency of anxious people to react histrioni- Finkel, Burnette, and Scissors (2007) exper- cally to a partner’s hurtful behaviors. Over- imentally enhanced attachment anxiety or all, research indicates that (1) people who measured weekly fluctuations in attachment are dispositionally secure can deal construc- anxiety for 6 months and found that height- tively with hurtful relational events, and (2) ened attachment anxiety reduced forgiveness contextual activation of the sense of security of a partner’s offenses. In addition, Han- can soften the typical maladaptive responses non, Rusbult, Finkel, and Kumashiro (2010) of dispositionally insecure people. found that, following an act of betrayal, the offending partner’s provision of a sense of Responses to Separations security to the injured partner (by genu- and Relationship Breakups inely expressing interest in being responsive Several studies have shown that attach- to the victim’s needs) promoted forgiveness ment orientations predict the intensity and and restoration of relational harmony. Kar- duration of distress following a romantic remans and Aarts (2007) found that sub- relationship breakup (e.g., Sbarra, 2006; liminal security priming (with the name Sbarra & Emery, 2005), divorce (Birn- of a loving other), compared with neutral baum, Orr, Mikulincer, & Florian, 1997), priming, elicited more automatic forgiveness wartime separation from a marital partner responses to interpersonal transgressions. (e.g., Medway, Davis, Cafferty, Chappell, & O’Hearn, 1995), or temporary separations Following this line of reasoning, Cassidy, from a romantic partner (e.g., Diamond, Shaver, Mikulincer, and Lavy (2009) showed Hicks, & Otter-H­ enderson, 2008). In these that experimentally heightening the sense of studies, distress intensification was a com- attachment security can reduce detrimental mon response of anxiously attached people, effects of dispositional attachment insecu- whereas attachment security was associated rities on cognitive and emotional reactions with faster emotional recovery and adjust- to hurtful experiences. Participants wrote a ment. For example, Sbarra (2006) collected description of an incident in which a close daily emotion data for 4 weeks from a sam- relationship partner criticized, disapproved, ple of young adults who had recently experi- rejected, or ostracized them. They then com- enced a romantic relationship breakup and pleted a computerized task in which they found that attachment anxiety was associ- were repeatedly exposed subliminally (for ated with slower recovery from sadness and 22 milliseconds on each occasion) to either anger. In a 21-day diary study, Diamond et a security-e­nhancing prime word (love, al. (2008) found that attachment anxiety was secure, affection) or a neutral prime word associated with sleeping problems, physical (lamp, staple, building). Immediately after symptoms, and higher levels of salivary cor- the priming trials, participants were asked tisol during and following days of physical to think again about the hurtful event they separation from a romantic partner brought had described and to rate how they would about by work-r­elated travel. Using fMRI react to such an event if it happened in the to observe brain responses to the recall of a future—­for example, how rejected they painful separation, Gillath, Bunge, Shaver, would feel, and how they would feel about Wendelken, and Mikulincer (2005) found themselves. that attachment anxiety was associated with higher activation of the left anterior tempo- In the neutral priming condition, the find- ral pole and left hippocampus (areas associ- ings confirmed the usual deactivating and ated with recall of sad thoughts) and lower hyperactivating strategies of avoidant and activation of the orbitofrontal cortex (an anxious people. Whereas avoidance scores area associated with emotional control). were associated with more defensive/hostile reactions, attachment anxiety was associ- ated with more intense feelings of rejec- tion, more crying, and more negative emo-

Attachment and Emotion Regulation 245 For avoidant individuals, the findings ductance during the task. However, Miku- depended on the nature of the separation lincer et al. (2004) found that avoidant peo- (Mikulincer & Shaver, 2007a). Avoidance ple’s ability to suppress separation-­related was associated with higher levels of distress thoughts was disrupted when a cognitive following divorce and wartime separations load—r­emembering a 7-digit number—w­ as but lower levels of distress and greater relief added to the experimental task. Under a high following temporary separations from, or cognitive load, avoidant individuals sud- permanent breakups with, dating part- denly evinced high availability of thoughts ners. It seems that avoidant people who of separation and negative self-t­raits. That can handle the distress of brief separations is, the suppressed material resurfaced in or the dissolution of casual bonds are less experience and behavior when a high cogni- successful in dealing with major separations tive demand was imposed. We suspect that requiring reorganization of relational rou- a similar resurfacing occurs when a high tines, goals, and plans. This fits with other emotional demand is imposed, as in the evidence, including that from experiments studies reviewed earlier that dealt with life-­ (e.g., Mikulincer, Dolev, & Shaver, 2004), threatening traumatic events (e.g., Berant et indicating that avoidant defenses collapse al., 2008). Recently, Ehrenthal, Friederich, under pressure. and Schauenburg (2011) found that avoidant attachment was associated with impaired There is also evidence that people with blood pressure recovery following recall of a different attachment orientations differ in painful separation, further emphasizing the the ways they cope with separation. For fragility of avoidant defenses. example, following divorce (Birnbaum et al., 1997) and temporary separations from a Reactions to the Death of a Close dating partner (Feeney, 1998), attachment-­ Relationship Partner anxious individuals were more likely to rely Few studies have directly examined associa- on emotion-f­ocused coping strategies, and tions between attachment orientations and avoidant individuals, on distancing strate- grief reactions following the death of a close gies. Similar coping strategies were noted by relationship partner. However, these stud- Davis, Shaver, and Vernon (2003) in a sur- ies have consistently found that attachment vey of more than 5,000 Internet respondents anxiety is associated with complicated grief who described romantic relationship break- reactions (e.g., Fraley & Bonanno, 2004; ups. Avoidant respondents were less likely Jerga, Shaver, & Wilkinson, 2011). For to seek support and more likely to cope example, Field and Sundin (2001) found that with the breakup alone, while avoiding new anxious attachment, assessed 10 months romantic involvements. Anxious respon- after the death of a spouse, predicted higher dents reacted with angry protests, height- levels of psychological distress 14, 25, and ened sexual attraction to the former partner, 60 months after the loss, and Fraley and intense preoccupation with the lost partner, Bonanno (2004) found that attachment a damaged sense of identity, and interference anxiety assessed 4 months after the loss of with school and work activities. In addition, a spouse predicted higher levels of anxiety, as a means of coping with separation, both depression, grief, trauma-­related symptoms, anxious and avoidant individuals used alco- and alcohol consumption 18 months follow- hol and drugs, which is not generally an ing the loss. effective coping strategy. With regard to avoidant attachment, Studies that induced thoughts about hypo- researchers have generally found no signifi- thetical or actual separations also provide cant association between this attachment important information about attachment-­ pattern and depression, grief, or distress related differences in regulating the distress following the death of a relationship part- resulting from these thoughts. For example, ner (e.g., Field & Sundin, 2001; Fraley & using a thought suppression task, Fraley Bonanno, 2004). However, Wayment and and Shaver (1997) showed that avoidant Vierthaler (2002) found that avoidance was people were highly effective in suppressing associated with increased somatic symp- separation-r­elated thoughts, as indicated tom levels following the death of a spouse. by less frequent thoughts of loss following the suppression task and lower skin con-

246 SOCIAL ASPECTS Recently, Jerga et al. (2011) found that and view themselves as superior, autono- avoidant attachment was positively associ- mous, and properly unemotional. This does ated with prolonged grief symptoms but not not keep them from reacting to relationship with typical or normative grief symptoms. partners with frustration, hostility, and That is, people who are generally avoidant denigration or from boosting their own self-­ in close relationships experience long-term esteem in the face of relationship losses by difficulties adjusting to the death of a rela- focusing disproportionately on their own tionship partner, even though they do not strengths and other people’s weaknesses. necessarily experience more intense typical grief symptoms. Attachment anxiety and attachment system hyperactivation are reactions to Jerga et al. (2011) also found that important relationships in which attach- relationship-­specific avoidance was nega- ment figures, often beginning with one or tively associated with both typical and both parents, reacted inconsistently to a prolonged grief symptoms. However, other person’s expressions of need, vulnerability, findings indicated that this association dis- and negative emotions, sometimes providing appeared when measures of relationship support but at other times being frustrating closeness and strength were statistically con- or inattentive (often because of their own trolled, suggesting that avoidant individuals self-f­ocused anxiety). This caregiver regi- may maintain relatively weak and emotion- men causes a person to believe that constant ally distant relationships with the deceased, vigilance, worry, and expressions of need, which in turn leaves them with less to grieve vulnerability, and retaliatory anger pay off, about. In other words, it may not be avoid- because they sometimes do capture a rela- ant attachment per se that protects avoidant tionship partner’s attention. Unfortunately, individuals from grief symptoms; it may be they can also alienate a person from his or the weakness of the emotional bonds they her initially favorable and loving relation- have to contend with when a relationship ship partners and evoke exactly what the partner dies. anxious person fears: rejection or abandon- ment. Thus, what began as a response to a Conclusions partial reinforcement schedule of attention In summarizing recent research on adult and support, and what became a pattern of attachment patterns and its implications noisy negativity, seems to the anxious per- for emotion regulation, we have shown son to lead to outcomes that confirm his or that attachment security is associated with her expectations and worst fears. appraisals and regulation efforts that are compatible with a balanced, open mind, These different patterns of emotional generally low levels of stress and distress, reaction and defense have been documented and constructive approaches to relationship in a remarkable variety of studies using maintenance. The two major dimensions experimental, interview, and observational of insecure attachment are associated with research methods. They are now being illu- characteristic strategies for dealing with minated further by neuroscientific stud- painful experiences in previous attachment ies. They provide strong specific support relationships. for Bowlby and Ainsworth’s attachment theory and its extension into the realm of Avoidance and attachment system deac- adult relationships (Mikulincer & Shaver, tivation are reactions to important relation- 2007a). However, more research is needed ships in which attachment figures, often if we are to understand better (1) the spe- beginning with one or both parents, reacted cific hyperactivating or deactivating strate- negatively to expressions of need, vulner- gies a particular insecure person will use in ability, and negative emotions. To cope particular situations, (2) the link between with that powerfully painful relationship attachment orientations and the relative influence, avoidant people have learned to balance of implicit and explicit emotion-­ downplay threats (i.e., trying not to appraise regulation attempts, and (3) the ways people events as threatening), suppress or deny feel- scoring high on both attachment anxiety ings of vulnerability and negative emotions, and avoidance regulate their emotions. In addition, future studies should attempt to integrate Bowlby’s (1988) notion that effec-

Attachment and Emotion Regulation 247 tive therapeutic interventions amount to 3. Sadness and depression. New York: Basic providing a secure base and help with emo- Books. tion regulation—pÂ

248 SOCIAL ASPECTS Davis, D., Shaver, P. R., & Vernon, M. L. (2003). theory and close relationships (pp. 189–219). Physical, emotional, and behavioral reactions New York: Guilford Press. to breaking up: The roles of gender, age, emo- Field, N. P., & Sundin, E. C. (2001). Attachment tional involvement, and attachment style. Per- style in adjustment to conjugal bereavement. sonality and Social Psychology Bulletin, 29, 871–884. Journal of Social and Personal Relationships, 18, 347–361. Dewitte, M., & De Houwer, J. (2008). Adult Finkel, E. J., Burnette, J. L., & Scissors, L. E. attachment and attention to positive and nega- (2007). Vengefully ever after: Destiny beliefs, tive emotional face expressions. Journal of state attachment anxiety, and forgiveness. Research in Personality, 42, 498–505. Journal of Personality and Social Psychology, Dewitte, M., De Houwer, J., Goubert, L., & 92, 871–886. Buysse, A. (2010). A multi-modal approach to Fraley, R., & Bonanno, G. A. (2004). Attachment the study of attachment-­related distress. Bio- and loss: A test of three competing models on logical Psychology, 85, 149–162. the association between attachment-r­elated avoidance and adaptation to bereavement. Dewitte, M., De Houwer, J., Koster, E. H. W., & Buysse, A. (2007). What’s in a name?: Personality and Social Psychology Bulletin, Attachment-­related attentional bias. Emotion, 30, 878–890. 7, 535–545. Fraley, R. C., & Shaver, P. R. (1997). Adult attachment and the suppression of unwanted Dewitte, M., Koster, E. H. W., De Houwer, J., thoughts. Journal of Personality and Social & Buysse, A. (2007). Attentive processing Psychology, 73, 1080–1091. of threat and adult attachment: A dot-probe Fraley, R. C., & Shaver, P. R. (2000). Adult study. Behaviour Research and Therapy, 45, romantic attachment: Theoretical develop- 1307–1317. ments, emerging controversies, and unan- swered questions. Review of General Psychol- Diamond, L. M., Hicks, A. M., & Otter-­ ogy, 4, 132–154. Henderson, K. A. (2006). Physiological evi- Gaines, S. O. Jr., Reis, H. T., Summers, S., Rus- dence for repressive coping among avoidantly bult, C. E., Cox, C. L., Wexler, M. O., et al. attached adults. Journal of Social and Per- (1997). Impact of attachment style on reactions sonal Relationships, 23, 205–229. to accommodative dilemmas in close relation- ships. Personal Relationships, 4, 93–113. Diamond, L. M., Hicks, A. M., & Otter-­ Gillath, O., Bunge, S. A., Shaver, P. R., Wen- Henderson, K. A. (2008). Every time you go delken, C., & Mikulincer, M. (2005). away: Changes in affect, behavior, and physi- Attachment-­style differences in the ability to ology associated with travel-r­elated separa- suppress negative thoughts: Exploring the neu- tions from romantic partners. Journal of Per- ral correlates. NeuroImage, 28, 835–847. sonality and Social Psychology, 95, 385–403. Granqvist, P., Mikulincer, M., Gurwitz, V., & Shaver, P. R. (2012). Experimental findings Ditzen, B., Schmidt, S., Strauss, B., Nater, U. on God as an attachment figure: Normative M., Ehlert, U., & Heinrichs, M. (2008). processes and moderating effects of internal Adult attachment and social support inter- working models. Journal of Personality and act to reduce psychological but not cortisol Social Psychology, 103, 804–818. responses to stress. Journal of Psychosomatic Hannon, P. A., Rusbult, C. E., Finkel, E. J., Research, 64, 479–486. & Kumashiro, M. A. (2010). In the wake of betrayal: Perpetrator amends, victim forgive- Ehrenthal, J. C., Friederich, H. C., & Schauen- ness, and the resolution of betrayal incidents. burg, H. (2011). Separation recall: Psy- Personal Relationships, 17, 253–278. chophysiological response-p­atterns in an Hobdy, J., Hayslip, B., Kaminski, P. L., Crowley, attachment-­related short-term stressor. Stress B. J., Riggs, S., & York, C. (2007). The role of and Health, 27, 251–255. attachment style in coping with job loss and the empty nest in adulthood. International Eisenberger, N. I., Master, S. L., Inagaki, T. K., Taylor, S. E., Shirinyan, D., Lieberman, M. Journal of Aging and Human Development, D., et al. (2011). Attachment figures activate a 65, 335–371. safety signal-­related neural region and reduce Holmberg, D., Lomore, C. D., Takacs, T. A., & pain experience. Proceedings of the National Price, E. L. (2011). Adult attachment styles Academy of Sciences USA, 108, 11721–11726. Feeney, J. A. (1998). Adult attachment and relationship-­centered anxiety: Responses to physical and emotional distancing. In J. A. Simpson & W. S. Rholes (Eds.), Attachment

Attachment and Emotion Regulation 249 and stressor severity as moderators of the Personality and Social Psychology Bulletin, coping sequence. Personal Relationships, 18, 21, 406–414. 502–517. Mikulincer, M., & Florian, V. (1998). The rela- Jerga, C., Shaver, P. R., & Wilkinson, R. B. tionship between adult attachment styles and (2011). Attachment insecurities and identifica- emotional and cognitive reactions to stressful tion of at-risk individuals following the death events. In J. A. Simpson & W. S. Rholes (Eds.), of a loved one. Journal of Social and Personal Relationships, 28, 891–914. Attachment theory and close relationships Karremans, J. C., & Aarts, H. (2007). The role (pp. 143–165). New York: Guilford Press. of automaticity in the inclination to forgive Mikulincer, M., & Florian, V. (1999). Maternal–­ close others. Journal of Experimental Social fetal bonding, coping strategies, and mental Psychology, 43, 902–917. health during pregnancy: The contribution of Kidd, T., Hamer, M., & Steptoe, A. (2011). attachment style. Journal of Social and Clini- Examining the association between adult cal Psychology, 18, 255–276. attachment style and cortisol responses to Mikulincer, M., Gillath, O., & Shaver, P. R. acute stress. Psychoneuroendocrinology, 36, (2002). Activation of the attachment system 771–779. in adulthood: Threat-r­elated primes increase Lemche, E., Giampietro, V. P., Surguladze, S. the accessibility of mental representations of A., Amaro, E. J., Andrew, C. M., Williams, S. attachment figures. Journal of Personality and C., et al. (2006). Human attachment security Social Psychology, 83, 881–895. is mediated by the amygdala: Evidence from Mikulincer, M., Hirschberger, G., Nachmias, O., combined fMRI and psychophysiological mea- & Gillath, O. (2001). The affective component sures. Human Brain Mapping, 27, 623–635. of the secure base schema: Affective priming Lopez, F. G., & Gormley, B. (2002). Stability with representations of attachment security. and change in adult attachment style over the first-year college transition: Relations to Journal of Personality and Social Psychology, self-­confidence, coping, and distress patterns. 81, 305–321. Journal of Counseling Psychology, 49, 355– Mikulincer, M., & Shaver, P. R. (2001). Attach- 364. ment theory and intergroup bias: Evidence Master, S. L., Eisenberger, N. I., Taylor, S. E., that priming the secure base schema attenu- Naliboff, B. D., Shirinyan, D., & Lieberman, ates negative reactions to out-­groups. Jour- M. D. (2009). A picture’s worth: Partner pho- nal of Personality and Social Psychology, 81, tographs reduce experimentally induced pain. 97–115. Psychological Science, 20, 1316–1318. Mikulincer, M., & Shaver, P. R. (2007a). Attach- Maunder, R. G., Lancee, W. J., Nolan, R. P., Hunter, J. J., & Tannenbaum, D. (2006). The ment in adulthood: Structure, dynamics, and relationship of attachment insecurity to subjec- change. New York: Guilford Press. tive stress and autonomic function during stan- Mikulincer, M., & Shaver, P. R. (2007b). Boost- dardized acute stress in healthy adults. Journal ing attachment security to promote mental of Psychosomatic Research, 60, 283–290. health, prosocial values, and inter-group toler- Medway, F. J., Davis, K. E., Cafferty, T. P., Chap- ance. Psychological Inquiry, 18, 139–156. pell, K. D., & O’Hearn, R. E. (1995). Family Mikulincer, M., Shaver, P. R., Sapir-Lavid, Y., disruption and adult attachment correlates & Avihou-Kanza, N. (2009). What’s inside of spouse and child reactions to separation the minds of securely and insecurely attached and reunion due to Operation Desert Storm. people?: The secure-­base script and its associa- Journal of Social and Clinical Psychology, 14, tions with attachment-­style dimensions. Jour- 97–118. nal of Personality and Social Psychology, 97, Mikulincer, M., Dolev, T., & Shaver, P. R. 615 – 633. (2004). Attachment-­related strategies during Mikulincer, M., Shaver, P. R., & Slav, K. (2006). thought-­suppression: Ironic rebounds and vul- Attachment, mental representations of others, nerable self-­representations. Journal of Per- and gratitude and forgiveness in romantic rela- sonality and Social Psychology, 87, 940–956. tionships. In M. Mikulincer & G. S. Goodman Mikulincer, M., & Florian, V. (1995). Appraisal (Eds.), Dynamics of romantic love: Attach- of and coping with a real-life stressful situa- ment, caregiving, and sex (pp. 190–215). New tion: The contribution of attachment styles. York: Guilford Press. Park, L. E. (2010). Responses to self-t­hreat: Linking self and relational constructs with approach and avoidance motivation. Social

250 SOCIAL ASPECTS and Personality Psychology Compass, 4, 201– Selcuk, E., Zayas, V., Günaydin, G., Hazan, C., 221. & Kross, E. (2012). Mental representations of Perunovic, M., & Holmes, J. G. (2008). Auto- attachment figures facilitate recovery follow- matic accommodation: The role of personal- ing upsetting autobiographical memory recall. ity. Personal Relationships, 15, 57–70. Quirin, M., Gillath, O., Pruessner, J. C., & Egg- Journal of Personality and Social Psychology, ert, L. D. (2010). Adult attachment insecurity 103, 362–378. and hippocampal cell density. Social Cogni- Shaver, P. R., & Mikulincer, M. (2002). tive and Affective Neuroscience, 5, 39–47. Attachment-­related psychodynamics. Attach- Quirin, M., Pruessner, J. C., & Kuhl, J. (2008). ment and Human Development, 4, 133–161. HPA system regulation and adult attachment Smeets, T. (2010). Autonomic and hypothalamic–­ anxiety: Individual differences in reactive and pituitary–­adrenal stress resilience: Impact of awakening cortisol. Psychoneuroendocrinol- cardiac vagal tone. Biological Psychology, 84, ogy, 33, 581–590. 290 –295. Raskin, P. M., Kummel, P., & Bannister, T. Waters, H. S., & Waters, E. (2006). The attach- (1998). The relationship between coping ment working models concept: Among other styles, attachment, and career salience in part- things, we build scriptlike representations nered working women with children. Journal of secure base experiences. Attachment and of Career Assessment, 6, 403–416. Human Development, 8, 185–198. Rusbult, C. E., Verette, J., Whitney, G. A., Wayment, H. A., & Vierthaler, J. (2002). Attach- Slovik, L. F., & Lipkus, I. (1991). Accommo- ment style and bereavement reactions. Journal dation processes in close relationships: Theory of Loss and Trauma, 7, 129–149. and preliminary empirical evidence. Journal of Williams, N. L., & Riskind, J. H. (2004). Adult Personality and Social Psychology, 60, 53–78. romantic attachment and cognitive vulnerabil- Sbarra, D. A. (2006). Predicting the onset of emo- ities to anxiety and depression: Examining the tional recovery following nonmarital relation- interpersonal basis of vulnerability models. ship dissolution: Survival analyses of sadness Journal of Cognitive Psychotherapy, 18, 7–24. and anger. Personality and Social Psychology Yárnoz-Yaben, S. (2009). Forgiveness, attach- Bulletin, 32, 298–312. ment, and divorce. Journal of Divorce and Sbarra, D. A., & Emery, R. E. (2005). The emo- Remarriage, 50, 282–294. tional sequelae of nonmarital relationship Zhang, F., & Labouvie-V­ ief, G. (2004). Stabil- dissolution: Analysis of change and intrain- ity and fluctuation in adult attachment style dividual variability over time. Personal Rela- over a 6-year period. Attachment and Human tionships, 12, 213–232. Development, 6, 419–437. Scharfe, E., & Bartholomew, K. (1995). Accom- Zilcha-Mano, S., Mikulincer, M., & Shaver, P. modation and attachment representations in R. (2011). Pets as safe havens and secure bases: couples. Journal of Social and Personal Rela- The moderating role of pet attachment orien- tionships, 12, 389–401. tations. Journal of Research in Personality, 45, 345–357.

Chapter 16 Attitudes, Evaluation, and Emotion Regulation Christopher R. Jones Tabitha Kirkland William A. Cunningham Research and theory in the domain of atti- contemporary perspectives on attitudes view tudes and evaluation are relevant to emotion evaluations as the product of dynamic pro- due to their many overlaps and parallels. cesses that are shaped by the demands of the Most fundamentally, they share at their core immediate situation. Relatively stable men- the concept of valence. Attitudes encode the tal representations underlie the very earli- valence of an object: that is, goodness or est evaluative processing; however, in a way badness, desirability or undesirability, plea- that can be viewed as fundamentally self-­ sure or pain. Valence is similarly integral to regulatory (e.g., Carver & Scheier, 1982), the concept of emotion; some theories argue evaluations are sensitive to various goals and that all emotions, at their core, can be dis- standards for thoughts, feelings, and behav- tilled to whether they are positive or nega- ior. tive (e.g., Russell, 2003; Watson & Tellegen, 1985). Valence’s function is to guide appro- Regarding the volitional self-­regulation priate action. In the case of attitudes, that of evaluation, it will be uncontroversial functionality lies in the ability of valence to note that people are not merely pas- to provide a rapid, default assessment of an sive receivers of their psychological experi- object’s value; we need not constantly judge ences; valence may be not only experienced the world anew. but also assessed and modified. Though evaluation has some characteristics of cued Yet this tool that may even be lifesaving in responding grounded by preexisting struc- times of peril can also be a liability. Reflex- ture, people do not respond to the same cue ively acting in accordance with one’s feelings in an invariable fashion. One may generally is not always appropriate; indeed, the feeling dislike something (say, golf) but temper that itself may be deemed problematic. For exam- affective response, or even reverse it, when ple, prejudiced group attitudes and evalua- the situation demands. For example, when tions may conflict with social norms or one’s offered the opportunity to golf with a work values, leading to reconciliatory processes of supervisor, one might be motivated to deem- justification or suppression (Crandall, Eshle- phasize the object’s disliked features and man, & O’Brien, 2002). We also argue that emphasize its liked ones (e.g., defining golf evaluation entails a process of affect regula- more by its aesthetic value than by, say, its tion in and of itself. Even in the absence of early-m­ orning tee times), or reevaluate its a particular desire to change one’s evalua- features (e.g., considering how golf’s diffi- tion, regulation is occurring. Increasingly, culty might constitute a virtue rather than 251

252 SOCIAL ASPECTS an annoyance). Evaluations are often shaped to deemphasize them, focusing instead on to conform to appropriate action. As such, contextual factors that shape experiences of when considering the use of attitudes to liking and disliking. In summary, we use the guide behavior, the regulation of affect is term attitude to refer to a structural entity important. One type of evaluative regulation in memory, and the term evaluation to refer entails the inhibition and reshaping of “gut to an appraisal process informed by an atti- reactions” of liking in accordance with their tude that unfolds over time in a particular perceived appropriateness. Beyond this, we context. Thus, evaluation can be considered emphasize processes within even quite rapid a process of affect regulation in the moment, and spontaneous evaluations (those very gut fostering context appropriateness. Given our reactions) that evidence self-­regulatory flex- terminology, attitudes may also be subject to ibility in experienced object-­directed affect. self-r­ egulation, such as when one attempts to Relatedly, Gross and Barrett (2011) sug- change how much one generally likes some- gest the possibility that emotion generation thing. We primarily focus on the former. and regulation can be considered the same process: Information may be simultane- It is the interplay of attitudes and evalua- ously processed as affectively relevant and tion that provides true functionality. A rigid informed by top-down processes that take retrieval system would be insensitive to con- contextual cues (e.g., situational appropri- text and beholden to even limited past expe- ateness) into account. Similarly, Todd, Cun- rience. A fully constructed system would be ningham, Anderson, and Thompson (2012) slow, inefficient, and too unresponsive to have suggested that emotion can be simulta- lessons learned. Instead, we have both capa- neously shaped and regulated through early bilities. The transformation of attitudes into attentional biases. In making this argument, evaluations is a form of emotion regulation we hope to highlight the potential for greater in which situational factors can play a role integration between the literatures on affect even in immediate responses. We argue that regulation and attitudes, which as yet have an attitude is a triggered affective response stood largely separate. shaped into a contextually appropriate eval- uation via a process of iterative reprocessing At the outset, some clarification of our (Cunningham & Zelazo, 2007). This updat- usage of the terms attitude and evaluation ing occurs as the situation unfolds; thus, will be useful. Some prominent definitions affect regulation is critical to all stages of have defined attitudes generally: predisposi- the attitude-­to-­evaluation transformation—­ tions to respond positively or negatively to from the processes typically considered an object (e.g., Allport, 1935; Eagly & Chai- automatic to the processes that are typically ken, 1993), or as a constellation of affec- considered controlled. Thus, although the tive, cognitive, and behavioral responses effortful inhibition of evoked attitudes is a associated with an object (Zanna & Rem- major topic, our conceptualization of regu- pel, 1988). In contrast, in a more specific lation goes beyond this to include a variety and limited definition, Fazio (2007) defines of processes that shape adaptive responding. attitudes as associations in memory between This broader view of emotion regulation objects and summated valenced information is consistent with evolving views on self-­ about those objects. This approach is the regulation generally, in which inhibition is one we take here, defining attitudes as repre- not a sine qua non (e.g., Fujita, 2011). sentational, which means that attitudes sig- nify valenced information stored in memory In this chapter, we review contemporary that has the potential to be activated. We do, attitude theories that entail an emphasis however, distinguish representational atti- on regulation. Following a discussion of tudes from evaluation—­a process by which two foundational models to most recent valence is constructed in both bottom-­up approaches, we focus in detail on contem- and top-down fashion. In this way, we inte- porary treatments of attitude and evaluation grate a representational definition, with the that illustrate the self-­regulatory shaping approach to attitudes typically considered of rapid response. We highlight the itera- most antithetical to it: the constructivist tive reprocessing (IR) model (Cunningham approach (e.g., Schwarz & Bohner, 2001). & Zelazo, 2007), which not only addresses This approach tends not to deny the exis- such processes descriptively but also provides tence of mental representations but chooses a perspective from neuroscience regarding their underlying brain substrates. We use the

Attitudes and Emotion Regulation 253 IR model as a framework to address differ- between implicit and explicit attitudes or ences in theory and empirical controversy evaluations is a common way of doing so. that characterize this literature. However, something like this distinction has been long recognized. For example, Hov- Two Foundational Attitude Models land, Janis, and Kelley (1953, cited in Petty Models of attitudes commonly empha- & Briñol, 2006, p. 740) distinguished atti- size some distinction between implicit and tudes (“implicit responses” that are “some- explicit attitudes or evaluative processes. times unconscious”) from opinions (“verbal Undoubtedly, this is in large part due to answers that one covertly expresses”). The the introduction of reaction-t­ime-based recent accumulated evidence from implicit implicit attitude measures, such as evalu- measures can be understood as validation ative priming (Fazio, Jackson, Dunton, of the long-­appreciated observation that & Williams, 1995) and the Implicit Asso- rapid, unintentional responses (what one ciation Test (IAT; Greenwald, McGhee, & might call “gut reactions”) can differ a great Schwarz, 1998). Researchers had long hoped deal from carefully considered evaluations. to find a solution to the problem of dishon- Accounting fully for the processes that dis- est responding to attitude measures out of tinguish them is a central issue in contem- social desirability concerns. These implicit porary attitude theory and obviously impor- measures offered the best solution to date by tant for self-r­ egulation. We first discuss two engaging participants in a task that might influential models that have been recently reveal attitudes without actually calling on built upon to explain differences between participants to express them. immediate and subsequent responses. Interestingly, the new implicit attitude The MODE Model measures appear to diverge from explicit The acronym MODE refers to Motiva- measures (self-r­eport) particularly when tion and Opportunity as DEterminants of participants’ self-­presentational concerns whether the attitude-­to-b­ehavior process are high (e.g., Fazio et al., 1995; Olson, is relatively spontaneous or deliberative in Fazio, & Hermann, 2007), yet the mea- nature. The MODE model (Fazio, 1990; sures also diverge even when participants’ Olson & Fazio, 2009) posits a single atti- incentives to misrepresent themselves are tude representation, and defines an attitude minimal. For example, Karpinski and Hil- as the association in memory between an ton (2001) found null relations between object and a summary evaluation thereof (see IAT and self-­report measures of preference Fazio, 2007). The model is largely agnostic for insects versus flowers and apples versus about the nature of attitude representation. candy bars, seemingly noncontroversial atti- It is readily understood in terms of a “schema tude objects. Furthermore, it appeared that plus tag” model, in which summary evalu- social influence (not necessarily intentional ations are a special form of tag associated misrepresentation) was especially evident on with an attitude object. However, it is also the implicit measure rather than the explicit consistent with more recent dynamic and measure. That is, the IAT suggested a greater connectionist models if the term summary affinity for the healthy (i.e., socially val- is understood to refer to distributed infor- ued) alternative than was evident in explicit mation in memory (Eiser, Fazio, Stafford, behavior. This example highlights the find- & Prescott, 2003). Regardless, the MODE ing that dissociations between implicit and model is primarily concerned with how and explicit measures are common and complex when attitudes determine behavior. (for a meta-­analytic review, see Hofmann, Gawronski, Gschwendner, Le, & Schmitt, The MODE model predicts whether the 2005), and that these dissociations cannot attitude-­to-b­ehavior process is relatively be attributed exclusively to misrepresenta- spontaneous or deliberative in nature. The tion on self-r­ eport measures or measurement model notes that the primary function of error and lack of structural fit (Payne, Burk- attitudes is to provide a rapid assessment ley, & Stokes, 2008). of objects in the environment to facilitate appropriate behavior in a timely fashion. Models need to account for these disso- Attitudes can be activated automatically, ciations, and incorporating some distinction and the likelihood of automatic attitude

254 SOCIAL ASPECTS activation is a function of the associative sensitive. The notion of an implicit attitude strength between the attitude object and the is often attributed to Greenwald and Banaji associated evaluation (see Fazio, 2001). If (1995), who wrote: “Implicit attitudes are there is little motivation for further consid- introspectively unidentified (or inaccurately eration of the object, for whatever reason, or identified) traces of past experience that low opportunity (e.g., time pressure or cog- mediate favorable or unfavorable feeling, nitive load) to do so, attitudes should guide thought, or action toward social objects” behavior in a spontaneous, largely auto- (p. 8). It was not until Wilson, Lindsey, and matic fashion, in which they promote con- Schooler’s (2000) dual-a­ttitudes model, gruent behaviors afforded by the situation. however, that a major process model of dual Controlled deliberation about the attitude attitudes was articulated. and behavior is likely only when the motiva- tion and opportunity to do so are present. The Dual‑Attitudes Model This deliberation might lead to the alteration In the original conception of “implicit atti- or rejection of an attitude and the behavior tudes,” such attitudes, by definition, were it facilitates, but it could also affirm it. In considered to be unknown to the evaluator fact, the model suggests that affirmation (Greenwald & Banaji, 1995). The absence is most typical, because the activated atti- of awareness is most directly the facet of tude fosters biased processing. Alteration or automaticity (see Bargh, 1994) referred to rejection of the attitude depends on various by the term implicit. However, the term has factors, including contextual influences on been used in a wide variety of ways, lead- the object’s utility, social influence, object ing some to argue that because implicit de reappraisal, and so on. facto has become a catchall descriptor of (relatively) automatic processes, it should be In summary, the MODE model posits a understood as such (De Houwer & Moors, single attitude associated with each discrete 2007). Wilson and colleagues (2000), for representation of an attitude object that may example, do not define implicit attitudes as be retrieved from memory, often automati- being characterized by unawareness. They cally. That attitude is likely to shape sub- suggested that “people are often aware, sequent judgment and behavior; however, at least fleetingly, of [implicit attitudes]” the attitude’s influence is especially likely (p. 105). What, then, is an implicit attitude to be direct for spontaneous behaviors. in this model? When the motivation and opportunity to deliberate are present, the attitude might According to the dual-a­ttitudes model be disregarded, adjusted for reasons includ- (Wilson et al., 2000), implicit attitudes are, ing context or apparent appropriateness, or in a word, old (or prior). That is, when an validated by deliberation. The MODE mod- attitude representation changes, it is not el’s implications for emotion regulation are fully supplanted by a new, altered attitude. straightforward. The model applies directly Rather, both representations can coexist in to circumstances in which one wishes to memory. Thus, an implicit attitude is a prior, control automatic emotional responses to a often deeply ingrained attitude. By virtue of particular object. It points to motivation and its precedence and having been rehearsed, opportunity as broad categories encompass- the implicit attitude is strongly associated ing many specific variables and emphasizes with the attitude object. Thus, it will likely that any major impediment to motivation or be automatically activated upon encounter- ability drastically reduces the likelihood of ing an attitude object. Similar to the MODE deliberative regulation. model, the dual-a­ ttitudes model posits that “capacity and motivation” are required to To account for discrepancies between retrieve the newer, explicit attitude effort- implicit and explicit attitudes, the MODE fully. Retrieval of the explicit attitude is perspective suggests that implicit measures expected to lead to its application, especially largely reflect automatically activated atti- for controlled behaviors. The simultaneous tudes, while explicit measures capture delib- activation of implicit and explicit attitudes erative evaluations, which may or may not may also demonstrate unintended influ- reflect attitudes for the reasons described. ence of the implicit attitude, especially on Another way of explaining these differences unmonitored or uncontrollable nonverbal is to evoke two different attitude representa- tions to which the measures are differentially

Attitudes and Emotion Regulation 255 responses (e.g., Dovidio, Kawakami, John- attitudes that were largely unregulated. son, Johnson, & Howard, 1997). Schema-plus-tag or implicit-a­ ttitude-­as-­ habitual-r­esponse conceptualizations of Wilson et al. (2000) further distinguish attitude share the consequence of rapid a typology of four kinds of implicit atti- responses that are quite inflexible and tudes, which have different characteristics context-f­ree. Affective regulation is then due to their differing origins. Briefly, an possible if and only if one is motivated and implicit attitude might arise from repres- able to suppress, adjust, enhance, or replace sion, in which an anxiety-­provoking implicit (i.e., effortfully retrieve an explicit atti- attitude is expelled from consciousness tude) an evaluation. However, the strong and is thus implicit in the sense of being form of this view is inconsistent with find- unknown. Another implicit attitude arises ings that are indicative of self-r­egulatory from motivated overriding, when an attitude responding even in rapid responses evident is unwanted and effortfully suppressed. A on implicit measures. For example, Lowery, third implicit attitude arises from automatic Hardin, and Sinclair (2001) demonstrated overriding, when the process of overrid- that racial attitudes as measured by an IAT ing such an implicit attitude itself becomes were sensitive to social tuning—­that is, they automatized through repetition. In this case, were influenced by the presumed attitudes capacity and motivation are not actually of other individuals present. Increasingly, required for the explicit rather than implicit recent models address adaptive flexibility in attitude to predominate. Finally, the dual-­ processes occurring during the construction attitudes model suggests that implicit atti- of an evaluation. Notably, a similar devel- tudes may differ from explicit attitudes sim- opment has occurred in theorizing about ply by virtue of having an entirely separate stereotyping. Automatic stereotype activa- neurocognitive substrate (i.e., independent tion was often viewed as inevitable (Bargh, systems; see Lieberman, 2007, for such a 1999; Devine, 1989) and could only be over- view) but does not elaborate much on this ridden by controlled processes. Subsequent possibility, which distinguishes some more research in this domain suggested a much recent models. more optimistic view of flexibility and self-­ regulation at the earliest stages of process- In some ways, the implications of the ing (e.g., Jones & Fazio, 2010; Moskowitz, dual-­attitudes model for emotion regulation Gollwitzer, Wasel, & Schaal, 1999; Sassen- are similar to those of the MODE model. berg & Moskowitz, 2005). In contemporary Certainly, the necessity of motivation and attitude models, we see a greater emphasis ability/capacity for any sort of controlled on malleability in rapid responses. This mal- emotion regulation is again highlighted. leability can be viewed as demonstrating the The different types of implicit attitudes fundamentally regulatory nature of evalua- described, however, are uniquely relevant. tion. We turn now to recent models includ- The case of repression, though its very exis- ing several variations on this theme. tence is controversial, poses an interesting problem. In this case, one’s own affective Contemporary Attitude Models reactions to an object are anxiety provok- The Associative–Propositional ing, instigating defensive reactions that sup- Evaluation Model press and exclude them from consciousness. Probably the most influential contemporary Motivated overriding essentially describes a approach to attitudes and evaluation is the phenomenon of successful affective regula- associative–p­ropositional evaluation (APE) tion, but it does not elaborate much about model, which originally was introduced how this occurs, especially under difficult to explain the complex literature on the circumstances, with the exception that impact of interventions that form or change automatizing a regulatory response may attitudes on implicit and explicit measures occur through repetition. This is an interest- (Gawronski & Bodenhausen, 2006). Since ing possibility in the realm of emotion regu- then, the APE model has been broadened to lation. address attitudes and evaluation more gener- ally (see Gawronski & Bodenhausen, 2011). Summary of Historical Perspectives To summarize, these models in their early incarnations posited implicit or activated

256 SOCIAL ASPECTS The APE model describes two interactive restoring balance and consistency have long but distinct types of processes that under- been recognized, such as rejecting one of the lie evaluation: associative and propositional propositions or introducing another propo- processes.1 Associative processes activate sition that reconciles or trivializes the appar- associated representations from memory. ent conflict. This activation is presumed to follow the principles of contiguity and similarity; con- Importantly, associative and proposi- tiguity refers to the spatiotemporal proxim- tional processes are argued to be interactive. ity of stimuli that determines the structure In addition to the bottom-­up processes in of memory, whereas similarity refers to the which activated associations determine the fit between encountered objects and men- kinds of propositions that are evoked, there tal representations. Propositional processes are also top-down processes that create involve the validation of information acti- or activate further associations (see Gaw- vated by associative processes. Unlike asso- ronski & Bodenhausen, 2011, for details). ciative processes, propositional processes The APE model is useful for explaining the entail subjective assessments of truth values highly varied patterns of results that have (i.e., veridicality of information). They are been observed when implicit and explicit posited to operate according to principles of measures follow manipulations meant to logical consistency, though in the sense of form or change attitudes. Earlier models, perceived coherence rather than formal logic particularly those that emphasize the sta- per se. Associative processes are argued to bility of implicit evaluations (e.g., Wilson underlie implicit evaluations, and propo- et al., 2000), have trouble explaining cir- sitional processes are argued to underlie cumstances in which implicit measures are explicit evaluations. Thus, if implicit evalua- more sensitive to manipulation than explicit tions are perceived as valid, they will be used measures (e.g., Gawronski & LeBel, 2008; for explicit evaluations; if they are not, they Karpinski & Hilton, 2001). The APE model, are modified. in contrast, can explain when information will recruit associative versus propositional The APE model suggests that, following processes. connectionist models of associative memory, information associated with a given atti- Some types of interventions are likely to tude object is distributed across a network have direct effects on associative processes of weighted connections. The activation (Gawronski & Bodenhausen, 2011). For of a given piece of information associated example, evaluative conditioning, a phenom- with an object is typically probabilistic, in enon in which an attitude toward an object a manner that maximizes the likelihood of forms or changes due to its co-o­ ccurrence the most relevant information being acti- with one or more valenced objects, is vated, which can be viewed as regulatory. likely to have these types of direct associa- The particularities of the input stimulus, as tive effects (for reviews, see De Houwer, well as incidentally active information due Thomas, & Baeyens, 2001; Jones, Olson, & to recency of use, determine which subset of Fazio, 2010). The repeated pairing of objects associations is activated. This allows for dif- during conditioning procedures facilitates ferent implicit evaluations as a function of association formation and strengthening context, consistent with malleability often for obvious reasons. Other types of inter- observed on implicit attitude measures (e.g., ventions are likely to have direct effects Barden, Maddux, Petty, & Brewer, 2004). on propositional processes. Inductions of These implicit evaluations provide the grist cognitive inconsistency (e.g., Gawronski for propositional processing. If the propo- & Strack, 2004) impact propositional pro- sitional implications of an implicit evalu- cesses directly but may leave relevant asso- ation are inconsistent with other salient ciations intact. Events might also directly propositions, an aversive state of cognitive influence both associative and propositional dissonance is evoked. Cognitive dissonance, processes. For example, a persuasive mes- induced by the simultaneous activation of sage could change the propositions that are conflicting information, is an aversive state considered and also create new associations of arousal that motivates efforts to allevi- in memory. ate it (Festinger, 1957). Various methods of Finally, events might directly influence one process and indirectly influence the

Attitudes and Emotion Regulation 257 other. Any change in associative structure opposed to a negative strategy of suppres- can potentially influence the propositions sion. However, other work suggests that that are generated and assessed, the path generating or encountering negations may taken to reduce cognitive dissonance, and so have positive influences on affect regulation. forth. The propositions an individual con- For example, Herbert, Deutsch, Sütterlin, siders can create or change the associative Kübler, and Pauli (2011; see also Mauss, strength of associations, even in the absence Evers, Wilhelm, & Gross, 2006) examined of external stimulation. A fuller account of startle eyeblink responses to pleasant and the details of the APE model (see Gawronski unpleasant nouns that had or had not been & Bodenhausen, 2006, 2011) is beyond the negated. They found (with ample processing scope of this chapter, but it should be noted time) online effects of negation consistent that the APE model offers many specific with the logical meaning: reduced startle hypotheses about the eliciting circumstances response to negated unpleasant words and and processes underlying patterns of atti- increased response to negated pleasant tude change evident in implicit and explicit words. Thus, the role of negation in affect measures. regulation remains controversial. One issue addressed by the APE model The APE model suggests that the inter- that has interesting implications for affect play of associative and propositional pro- regulation concerns negation. The act of cesses underlies evaluation. It is sometimes negating a proposition may be insufficient misunderstood to be a dual-s­ystem or to change the relevant associative struc- dual-r­epresentation model but it explicitly ture in memory. Indeed, though negation is not (Gawronski & Bodenhausen, 2011, may influence propositional processes in pp. 104–105). The APE model argues that a straightforward manner (i.e., processing all information is stored in the form of asso- consistent with the negated proposition), the ciations, and that there is not a separate act of negation may have ironic effects at store for propositions, thus rejecting the the associative level due to the coactivation possibility of dual representations. The APE of elements (e.g., Deutsch, Gawronski, & model does not explicitly endorse or reject Strack, 2006). For example, telling oneself, a dual-­system approach, which emphasizes “I am not afraid of giving a speech,” may separate mental systems (i.e., neurocognitive have some immediate positive consequences substrates).3 Some researchers believe that due to its consciously processed implica- the sometimes stark dissociations between tion. However, this also risks increasing the observed implicit and explicit evaluations (automatic) association in memory between are best explained by their origination from fear and public speaking merely because of separate systems. We turn now to a recent their simultaneous activation. Gawronski model of dual systems to explore this idea in and Bodenhausen (2011) suggest that the greater detail. overall success of regulating future affec- tive reactions depends on whether unwanted The Systems of Evaluation Model propositions are negated (“I’m not afraid of The systems of evaluation model (SEM) public speaking”) or preferable propositions describes an associative system that pro- are affirmed (“Public speaking is fun”), duces implicit evaluations and a rule-based implying that affirmation works better than system that produces explicit evaluations negation for emotion regulation.2 For exam- (McConnell & Rydell, in press; Rydell & ple, a study in which stereotypical associ- McConnell, 2006). It builds on prior dual-­ ates were negated led to increased automatic system models of cognition, particularly stereotype activation and, significantly for Sloman’s (1996) model of fast and slow the question of affect regulation, increased learning systems of reasoning. These par- bias on an evaluative priming task (Gaw- tially independent systems operate in par- ronski, Deutsch, Mbirkou, Seibt, & Strack, allel and differ in the type of knowledge 2008). Gawronski and Bodenhausen (2011) they use and the operations conducted on suggested that such ironic effects may bear that knowledge. The associative system some responsibility for the general superi- reflects associations governed by principles ority in emotion regulation (Gross, 1998) of similarity and contiguity, whereas the of an affirmative strategy of reappraisal as

258 SOCIAL ASPECTS rule-based system operates on symbolic rep- positive followed by negative, or vice versa), resentations (e.g., language) that are sub- explicit attitude measures showed consider- jected to operations of logic and deductive able sensitivity to new information, while reasoning. Because implicit measures tend implicit measures were largely unaffected by to reflect construct activation and explicit it. When 100 counterattitudinal behaviors measures tend to reflect symbolic reason- followed the initial 100, both types of mea- ing, the measures differ in their sensitivity sures reflected the new information. This is to the output of the two systems. This sug- interpreted as being consistent with the pres- gests that the primary distinction between ence of a slow-l­earning associative system these systems is similar to the APE model. and fast rule-based system. In other work However, a major difference is that the SEM supportive of the SEM, a series of sublimi- emphasizes distinctive characteristics of the nal primes was paired with target individu- two systems that lead to differential rates of als (Rydell, McConnell, Mackie, & Strain, change. The SEM suggests that the explicit 2006). Because repeated subliminal priming system is faster, while the implicit system might produce the gradual association of is slower, both to form and to change atti- affect with an attitude object, it would be tudes. Explicit evaluations “can be formed expected to influence the associative sys- and modified relatively quickly because tem. Moreover, by virtue of being sublimi- logic and syllogism are responsive to one’s nal, such a manipulation would not provide deliberate goals and deductive reasoning grist for the rule-based system. However, processes,” whereas implicit evaluations the rule-based system should be especially “typically are slower to form and change sensitive to the consciously accessible behav- because they are based on accumulated atti- ioral information that was presented, as evi- tude object-­evaluation pairings in memory” denced by influences on explicit measures. (McConnell & Rydell, in press). In this way, Results in line with these hypotheses were the SEM bears some resemblance to the obtained: Across several studies, implicit earlier dual-­attitudes model of Wilson and attitudes formed and changed in response colleagues (2000), insofar as an implicit atti- to subliminal primes, and explicit attitudes tude tends to be distinguished by stability. formed and changed in response to con- However, the SEM does not posit separate sciously accessible information. stores for “old” and “new” attitudes, nor does it consider prior attitude change neces- An interesting aspect of the SEM with sarily responsible for divergence on implicit wide implications for affect regulation is the and explicit measures. Though implicit mea- study of the consequences of discrepancy sures are sometimes quite labile, this can be between implicit and explicit evaluations. reconciled with the SEM by noting that this One consequence of implicit–e­xplicit dis- often appears to be due to activation of dif- crepancy involves the extent of information ferent subset of associations from memory processing. It has been suggested that these that influence an implicit measure, among discrepancies lead to increased information other reasons, rather than a rapid change in processing (e.g., Petty, Tormala, Briñol, & particular associative representations. Jarvis, 2006). Moreover, recent work has suggested that this may be due to the arousal The SEM is supported by research dem- of cognitive dissonance caused by the con- onstrating strong dissociations between flicting outputs of the two systems (Rydell, implicit and explicit measures, consistent McConnell, & Mackie, 2008). Seeking out with a dual-­systems view. For example, information to resolve the discrepancy is a Rydell and McConnell (2006, Experiment major route to dissonance reduction. How- 2) presented participants with a series of ever, the source of these aversive feelings behaviors performed by a target individual. can be mysterious to those experiencing Presenting a large number of positive or neg- them, and consequently, the feelings might ative behaviors established positive or nega- be misattributed to inappropriate sources. tive attitudes, respectively, that were evident Specifically, negative arousal caused by atti- on both implicit and explicit measures. This tude formation with valence-­inconsistent was followed by a series of counterattitudinal information led participants to report lower behaviors. When 100 initial behaviors were subjective well-being (Rydell & Durso, followed by 20 conflicting behaviors (i.e., 2012). Thus, the SEM suggests that when

Attitudes and Emotion Regulation 259 the systems of evaluation produce discrep- association activation is acceptance, the pre- ant evaluations, negative affect and affective sumption of validity (see also Gilbert, 1991). regulation tend to follow. Another relevant Therefore, it tends to matter more when an consequence of implicit–e­ xplicit discrepancy invalidity tag is not retrieved, in which case a involves affective forecasting. It has also been rejected evaluation is likely to influence judg- suggested that thoughtful affective forecast- ment and behavior. The MCM distinguishes ing relies on explicit evaluations of objects, between explicit ambivalence, in which both ignoring implicit evaluations (McConnell, positive and negative evaluations are associ- Dunn, Austin, & Rawn, 2011). Consistent ated with validity tags, and implicit ambiva- with this idea, implicit, but not explicit, atti- lence, in which one association is validated tudes predicted error in affective forecasting and the other is not. Explicit ambivalence is when considering future enjoyment of foods. perceived as subjectively conflicting when it is experienced, but implicit ambivalence is The Meta‑Cognitive Model not, due to the conscious rejection of either Petty and colleagues’ meta-­cognitive model the positive or negative evaluation. (MCM; Petty, 2006; Petty et al., 2006; Petty, Briñol, & DeMarree, 2007) has elements of In summary, the MCM suggests that most of the other models we have discussed. the creation and retrieval of validity tags It focuses on how attitudes might be stored serve a self-­regulatory function by mark- in memory, with particular emphasis on ing which associations are (in)appropriate meta-c­ ognitive perceptions of the validity of for guiding action. Unlike the other models responses. Moreover, the MCM draws on discussed, the MCM particularly may be research arguing that the neural substrates seen as clearly concerning the regulation of of positive and negative valence are dis- attitudes per se. Validity tags concern the tinct (e.g., Cacioppo, Gardner, & Berntson, associative network of an attitude and the 1997), suggesting that separate positive and adaptive regulation of relatively general- negative evaluations of an object might be ized responses that can guide later action. represented in memory. Though all models concern the regulation of attitudes insofar as the process of evaluation According to the MCM, though many not only recruits but also changes attitudes attitudes are predominantly univalent, it is in memory, the MCM is particularly use- not uncommon for an object to be associ- ful for considering how attitude structure ated with both positive and negative sum- changes over time. mary evaluations. Whether the positive or negative attitude associated with an object The Iterative Reprocessing Model is activated depends on associative strength, and Dynamism recency of prior positive and negative expe- Our final contemporary perspective empha- riences, and whether the current context sizes dynamism in evaluation and regula- is a better fit with the positive or negative tion. Dynamism refers to the notion that association. Reflecting on these attitudes mental representations are constantly evolv- can lead them to become associated with a ing states in which partial representations tag, marking them as either valid or invalid. are cascading, competing, and interactive. The associated confidence indicator serves The dynamical approach to processing (e.g., as a signal to utilize or disregard the atti- Spivey, 2007) has been applied to multiple tude with minimal reflection required. social psychological concepts, including However, the MCM notes that validity tags categorization (Freeman, Ambady, Rule, & also vary in their associative strength to the Johnson, 2008), stereotyping (Freeman & evaluation and in accessibility, meaning that Ambady, 2009), and evaluation (Wojnowicz, sometimes the tag may not be retrieved from Ferguson, Dale, & Spivey, 2009). Wojnow- memory when the attitude object is. Often, icz and colleagues argue that explicit evalu- attitudes marked with an invalidity tag are ations are “merely the end result of a com- prior assessments that have been changed plex, non-l­inear, time-d­ ependent process of but still may exert an influence (Petty et al., multiple less-­explicit attitudes competing 2006). The MCM concurs with the APE in with one another over hundreds of millisec- the assumption that the default response to onds” (p. 1428). In early processing, mul-

260 SOCIAL ASPECTS tiple patterns of evoked activity are partially construal of a situation, and because cate- consistent with various evaluations. Over gories and construals shape which attitude time, “a continuous accrual of information aspects are foregrounded and made more causes the distributed pattern to dynami- active, stimuli initiate an iterative sequence cally ‘sharpen’ into a confident (selected) of evaluative processes (the evaluative cycle) interpretation, forcing other, partially acti- through which the stimuli are interpreted vated, competing alternative[s] . . . to gradu- and reinterpreted in light of an increasingly ally die out” (p. 1429), an idea consistent rich set of contextually meaningful repre- with the connectionist modeling notion of sentations. Such a view blurs the traditional an “attractor state” (e.g., Conrey & Smith, distinctions found in dual-­systems or dual-­ 2007). Selection is spurred by a cyclical pro- process models of attitudes and attitude cessing loop between higher-o­ rder integra- regulation. Indeed, although evaluations tive brain regions and lower-level informa- that are based on few iterations of the evalu- tional sources, in which the former enact ative cycle may be thought to be relatively representation competition that promotes automatic or implicit, they can be shaped or inhibits alternatives. Wojnowicz and col- by higher-­order processes before they are leagues (2009) drew on research on racial ever encountered (Cunningham, Van Bavel, attitudes (e.g., Fazio et al., 1995) suggesting Arbuckle, Packer, & Waggoner, 2012; Todd that negative partial evaluations of black tar- et al., 2012). Furthermore, although evalua- gets would be common early in participants’ tions that are based on additional iterations processing streams but would be subsumed may become more integrated, they do not by more positive evaluations due to the moti- necessarily require or use new representa- vation to control prejudice. They anticipated tions to generate ongoing evaluations. that the dynamic evolution of an evaluation would be evident in subtle behavioral traces Given the iterative nature of the model, as participants moved a cursor to associ- evaluations are proposed to be the dynamic ate “black” with “like” or “dislike.” Their result of an integrated set of distributed pro- theory specifically predicted a curvature in cesses, each of which responds to and resolves trajectory toward the eventually unchosen specific computational problems (see Cun- alternative (“dislike”) when expressing lik- ningham & Johnson, 2007). Evaluation is ing for the target “black” that would not be an emergent property of multiple processes evident when the target was “white.” This is that unfold over time. Critically, the model what they observed. proposes that a common set of processes is consistently involved in generating current The IR model (Cunningham & Zelazo, evaluations that are shaped by both bottom- 2007) considers such dynamism central in u­ p and top-down influences. Thus, evalua- regulating evaluative responses. As noted, tive processes are part of an iterative cycle: the terms attitude and evaluation are often With every iteration, the current evaluation used interchangeably, but we believe that of a stimulus is adjusted in light of addi- it is useful to refer to them as qualitatively tional contextual and motivational informa- different. Whereas an attitude is a relatively tion in order to create an updated evalua- stable set of representations of a stimulus, an tion. Information is continually passed back evaluation reflects one’s current appraisal of from relatively higher-­order to relatively the stimulus. Evaluative processes transform lower-order processes, and the evaluation is attitudes into evaluations. In other words, to recalculated. This “reseeding” of informa- generate an evaluation, one uses preexisting tion allows for the foregrounding of relevant attitudes to retrieve useful affective informa- (and backgrounding of irrelevant) attitude tion about a stimulus, but the evaluation also representations and contextual information takes into consideration information about in order to develop incorporate current goals the environment and context, as well as cur- and standards, and allows for the regulation rent goals (see Figure 16.1). Furthermore, of an evaluation response to come into line because information is distributed, not all with situational or motivational constraints. aspects of the attitude are activated; as such, At each iteration, the current evaluation only the currently active aspects of attitude serves as input for ongoing evaluative pro- can shape an evaluation. Critically, because cessing; as such, earlier evaluations are likely active representations shape perception and to bias subsequent evaluative processing by

Attitudes and Emotion Regulation 261 Attitude Representations IterationT Stimulus Evaluative Evaluation1 Construal1 Processes Stimulus IterationT+~200ms Time Stimulus Stimulus Evaluative Evaluation2 Construal2 Processes IterationT+~400ms Stimulus Evaluation3 Stimulus Construal3 Evaluative Processes Iterationn FIGURE 16.1.  The iterative reprocessing model. From Cunningham, Zelazo, Packer, and Van Bavel (2007). Copyright 2007 by The Guilford Press. Reprinted by permission. influencing attention, information seeking, Thus, unlike models that propose a stark stimulus construal, and so forth. Impor- difference between automatic and controlled tantly, while conscious deliberation exerts processes, the IR model suggests a contin- an influence on evaluative processing, infor- uum from relatively automatic to controlled mation about the valence and arousal value evaluative processes that can operate on a set of a stimulus continues to be represented in of representations. At the early stages of pro- subcortical structures. cessing, the strongest weights associated with an attitude give rise to a specific pattern of To understand when people are more activation, and result in quick and automatic likely to generate complex evaluations, the IR evaluations. However, with more iterations model proposes that two competing motiva- and the potential for reflective processing tional drives influence the extent of evalua- in areas of the brain responsible for higher tive processing. First, a drive to minimize the processing (e.g., prefrontal cortex), evalua- discrepancy between one’s evaluation and tions are shaped by a dynamic interaction of the hedonic environment (i.e., to minimize several bottom-u­ p and top-down processes. error) increases reflective processing dur- This interaction allows for the foreground- ing evaluation. Second, a drive to minimize ing and backgrounding of particular pat- processing demands decreases in reflective terns of activation in accordance with cur- processing during evaluation. These oppos- rent contexts and goals. Evaluations based ing drives create a dynamic tension that can on additional iterations are generally more help individuals to strike a delicate balance reflective. As individuals engage in reflec- between an initial “gut” response and evalu- tive reprocessing, they are able to formulate ations that are more nuanced but not com- more complex, nuanced representations of putationally overwhelming. The influence of a stimulus (e.g., allowing a stimulus to be these competing motivations likely varies as understood in terms of multiple conflicting a function of situational demands, current dimensions of evaluation). This hierarchical goals, and individual differences in process- approach, which views reflection as a matter ing style.

262 SOCIAL ASPECTS of degree, is consistent with contemporary delivery system rather than a tasty snack, characterizations of prefrontal cortical func- the nature of active representations will be tion (Bunge & Zelazo, 2006). more negative, and a more negative evalu- ation will be constructed. This process But although it is possible to draw parallels is central to emotion regulation, because between standard dual-p­ rocess models and changing the value of the goal state or the the IR model, the nature of the relationship experienced outcome is the most effective between what can be considered automatic means to feel better (e.g., “I didn’t actually and controlled differs in important ways. want that donut, because it is bad for me”). The IR model proposes that what is typically Furthermore, attitudes and their transfor- considered more controlled or reflective pro- mation into information are essential for all cessing in dual-­process models merely biases stages of emotion regulation, from situation which representations remain active and the selection (one can use attitudes to determine complexity of the evaluation that is possible whether a situation is likely to be good or given this set of active representations. Thus, bad) to attention deployment (attitudes the biasing of representations can come prior guide attention; Roskos-E­ woldsen & Fazio, to processing any given stimulus, allowing 1992). Attitudes provide the expectations for automatic regulation—­the initial rep- for a stimulus, and this information is criti- resentation, categorization, and evaluation cal to guiding responses that lead to good of an object can be modified at the earliest outcomes and more positive emotions. levels of processing. Indeed, attention to one category (race) or another (age) for stimuli Conclusion that can be multiply categorized leads to We hope the reader will find conceptual greater affective priming to the focal cat- parallels within the models reviewed and egory (Gawronski, Cunningham, LeBel, & emotion regulation. Many of the automatic Deutsch, 2010). That is, foregrounding one influences of goals and standards on auto- category or the other changed the initial matic evaluation, and the means and eliciting evaluation as measured by affective priming. circumstances of effortful self-r­egulation of attitudes and evaluation, are likely relevant Linking the Literatures on Attitudes, to the regulation of emotion. Unfortunately, Evaluation, and Emotion Regulation these literatures have as yet seen little inte- In this chapter, we suggest that the pro- gration despite their potential to inform one cesses by which attitudes are transformed another. For example, attitude theory has into evaluations may be more dynamic that often neglected the role of arousal, which previously considered. As such, the classical has been integral to the study of emotion. distinctions between automatic/implicit and We end with a final note on the applicability controlled/explicit processes may need to of models of evaluation for emotion regula- be modified to articulate the dynamic and tion. Although typically studied in isolation iterative nature of evaluative processes. In from one another, the processes involved in doing so, many parallels with the literature attitudes greatly overlap with those involved on affect and its regulation are noted. For in emotion regulation. Indeed, as suggested example, to the extent that evaluations are in this chapter, the critical question con- constantly updated, and regulation (defined cerning how one converts attitudes (stored as biasing representations) exists through- representations about the world) into cur- out the process, this suggests that the ways rent evaluations (temporary affective experi- that we shape affect will be important for ences that can be used to drive thoughts and the types of representations that are active, behavior) can be viewed fundamentally as a and how we use them to generate appropri- regulatory process in which an evoked affec- ate evaluations. For example, the processes tive response is shaped to reflect the exigen- of appraisal and reappraisal will be essential cies of the moment. for determining the meaning of an object in a situation, and determining its value. To This view goes beyond more traditional the extent that one is a successful dieter, by notions of regulation, which typically involve reappraising a donut as an efficient calorie

Attitudes and Emotion Regulation 263 the inhibition or alteration of an automatic References response, dependent upon conscious will and ability. We argue that the translation of Allport, G. W. (1935). Attitudes. In C. Murchi- representation (attitude) to experience (eval- son (Ed.), Handbook of social psychology. uation) involves regulation, as active repre- Worcester, MA: Clark University Press. sentations are shaped and reshaped across multiple levels of organization. At lower lev- Barden, J., Maddux, W. W., Petty, R. E., & els, the perceptual system helps to regulate Brewer, M. B. (2004). Contextual modera- what stimuli to attend to and what stimuli to tion of racial bias: The impact of social roles ignore by biasing attention, while at higher on controlled and automatically activated atti- levels, the more reflective regulatory systems tudes. Journal of Personality and Social Psy- make use of similar foregrounding and back- chology, 87, 5–22. grounding processes. As such, the ability to regulate one’s emotions effectively has been Bargh, J. A. (1994). The four horsemen of auto- linked to the ways in which these lower- and maticity: Awareness, intention, efficiency, and higher-o­ rder processes interact (Lee, Heller, control in social cognition. In R. S. Wyer, Jr. van Reekum, Nelson, & Davidson, 2012). & T. K. Srull (Eds.), Handbook of social cog- Critically, the ability to reinterpret or recon- nition (Vol. 1, 2nd ed., pp. 1–40). Hillsdale, strue incoming information flexibly—o­r NJ: Erlbaum. cognitive flexibility (Scott, 1962)—is an essential feature of these models and may Bargh, J. A. (1999). The cognitive monster: The be vital for emotion regulation. Although case against controllability of automatic ste- dynamic perspectives on attitudes are only reotyping effects. In S. Chaiken & Y. Trope beginning to receive attention, it is our hope (Eds.), Dual process theories in social psy- that these perspectives will be influential in chology (pp. 361–382). New York: Guilford guiding future research beyond the scope of Press. attitudes alone. After all, these models, at their core, are models of cognitive process- Bunge, S. A., & Zelazo, P. D. (2006). A brain- ing, and as such are relevant to and overlap based account of the development of rule use with many domains, emotion regulation in childhood. Current Directions in Psycho- included. logical Science, 15, 118–121. Notes Cacioppo, J. T., Gardner, W. L., & Berntson, G. G. (1997). Beyond bipolar conceptualizations 1. This basic distinction between associative and and measures: The case of attitudes and evalu- propositional processes is shared with Strack ative space. Personality and Social Psychology and Deutsch’s (2004) reflexive–­impulsive Review, 1, 3–25. model (RIM), which in turn drew from mul- tiple earlier dual-­process theories (see Smith Carver, C. S., & Scheier, M. F. (1982). Control & DeCoster, 2000, for a review). However, theory: A useful conceptual framework for the APE model focuses on evaluation per personality—S­ ocial, clinical, and health psy- se, whereas the RIM is explicitly a model of chology. Psychological Bulletin, 92, 111–135. behavior. Conrey, F. R., & Smith, E. R. (2007). Attitude 2. It should be noted that some negations representation: Attitudes as patterns in a dis- are familiar (“no way”) and/or very easily tributed, connectionist representational sys- reversed (“not good”), such that their appro- tem. Social Cognition, 25, 718–735. priate meaning may be encoded even when cognitive resources are strained (e.g., Mayo, Crandall, C. S., Eshleman, A., & O’Brien, L. Schul, & Burnstein, 2004), rendering ironic (2002). Social norms and the expression and effects unlikely in these cases. suppression of prejudice: The struggle for internalization. Journal of Personality and 3. However, its authors express skepticism that Social Psychology, 82, 359–378. a strong dual-s­ystems approach is consistent with emerging evidence from neuroscience. Crone, E. A., & Van der Molen, M. W. (2004). Developmental changes in real-life decision-­ making: Performance on a gambling task pre- viously shown to depend on the ventromedial prefrontal cortex. Developmental Neuropsy- chology, 25, 251–279. Cunningham, W. A., & Johnson, M. K. (2007). Attitudes and evaluation: Toward a component process framework. In E. Harmon-Jones & P.

264 SOCIAL ASPECTS Winkielman (Eds.), Social neuroscience: Inte- ogy of attitudes. Fort Worth, TX: Harcourt grating biological and psychological explana- Brace Jovanovich. tions of social behavior (pp. 227–245). New Eiser, J. R., Fazio, R. H., Stafford, T., & Prescott, York: Guilford Press. T. J. (2003). Connectionist simulation of atti- Cunningham, W. A., Johnson, M. K., Raye, C. tude learning: Asymmetries in the acquisition L., Gatenby, J. C., Gore, J. C., & Banaji, M. of positive and negative evaluations. Personal- R. (2004). Separable neural components in the ity and Social Psychology Bulletin, 29, 1221– processing of Black and White faces. Psycho- 1235. logical Science, 15, 806–813. Fazio, R. H. (1990). Multiple processes by which Cunningham, W. A., Raye, C. L., & Johnson, attitudes guide behavior: The MODE model M. K. (2004). Implicit and explicit evalua- as an integrative framework. In M. P. Zanna tion: fMRI correlates of valence, emotional (Ed.), Advances in experimental social psy- intensity, and control in the processing of atti- chology (Vol. 23, pp. 75–109). New York: tudes. Journal of Cognitive Neuroscience, 16, Academic Press. 1717–1729. Fazio, R. H. (2001). On the automatic activation Cunningham, W. A., Van Bavel, J. J., Arbuckle, of associated evaluations: An overview. Cog- N. L., Packer, D. J., & Waggoner, A. S. (2012). nition and Emotion, 15, 115–141. Rapid social perception is flexible: Approach Fazio, R. H. (2007). Attitudes as object-­ and avoidance motivational states shape P100 evaluation associations of varying strength. responses to other-race faces. Frontiers in Social Cognition, 25, 603–637. Human Neuroscience, 6, 140. Fazio, R. H., Jackson, J. R., Dunton, B. C., & Cunningham, W. A., & Zelazo, P. D. (2007). Williams, C. J. (1995). Variability in auto- Attitudes and evaluations: A social cognitive matic activation as an unobtrusive measure neuroscience perspective. Trends in Cognitive of racial attitudes: A bona fide pipeline? Jour- Sciences, 11, 97–104. nal of Personality and Social Psychology, 69, Cunningham, W. A., Zelazo, P. D., Packer, D. 1013–1027. J., & Van Bavel, J. J. (2007). The Iterative Festinger, L. (1957). A theory of cognitive dis- Reprocessing Model: A multilevel framework sonance. Evanston, IL: Row Peterson. for attitudes and evaluation. Social Cognition, Freeman, J. B., & Ambady, N. (2009). Motions 25, 736–760. of the hand expose the partial and parallel De Houwer, J., & Moors, A. (2007). How to activation of stereotypes. Psychological Sci- define and examine the implicitness of implicit ence, 20, 1183–1188. measures. In B. Wittenbrink & N. Schwarz Freeman, J. B., Ambady, N., Rule, N. O., & (Eds.), Implicit measures of attitudes: Proce- Johnson, K. L. (2008). Will a category cue dures and controversies (pp. 179–194). New attract you?: Motor output reveals dynamic York: Guilford Press. competition across person construal. Journal De Houwer, J., Thomas, S., & Baeyens, F. (2001). of Experimental Psychology: General, 137, Associative learning of likes and dislikes: A 673 – 690. review of 25 years of research on human eval- Fujita, K. (2011). On conceptualizing self-­ uative conditioning. Psychological Bulletin, control as more than the effortful inhibition of 127, 853–869. impulses. Personality and Social Psychology Deutsch, R., Gawronski, B., & Strack, F. (2006). Review, 15, 352–366. At the boundaries of automaticity: Negation Gawronski, B., & Bodenhausen, G. V. (2006). as reflective operation. Journal of Personality Associative and propositional processes in and Social Psychology, 91, 385–405. evaluation: An integrative review of implicit Devine, P. G. (1989). Stereotypes and prejudice: and explicit attitude change. Psychological Their automatic and controlled components. Bulletin, 132, 692–731. Journal of Personality and Social Psychology, Gawronski, B., & Bodenhausen, G. V. (2011). 56, 5–18. The associative–p­ropositional evaluation Dovidio, J. F., Kawakami, K., Johnson, C., John- model: Theory, evidence, and open ques- son, B., & Howard, A. (1997). On the nature tions.  Advances in Experimental Social Psy- of prejudice: Automatic and controlled pro- chology, 44, 59–127. cesses. Journal of Experimental Social Psy- Gawronski, B., Cunningham, W. A., Lebel, E. chology, 33, 510–540. P., & Deutsch, R. (2010). Attentional influ- Eagly, A. H., & Chaiken, S. (1993). The psychol- ences on affective priming: Does categoriza-

Attitudes and Emotion Regulation 265 tion influence spontaneous evaluations of effects on weapon identification. Personality multiply categorizable objects? Cognition, 24, and Social Psychology Bulletin, 36, 1073– 1008–1026. 1085. Gawronski, B., Deutsch, R., Mbirkou, S., Seibt, Jones, C. R., Olson, M. A., & Fazio, R. H. B., & Strack, F. (2008). When “just say no” (2010). Evaluative conditioning: The “how” is not enough: Affirmation versus negation question. In M. P. Zanna & J. M. Olson training and the reduction of automatic ste- (Eds.), Advances in experimental social psy- reotype activation. Journal of Experimental chology (Vol. 43, pp. 205–255). New York: Social Psychology, 44, 370–377. Academic Press. Gawronski, B., & LeBel, E. P. (2008). Under- Karpinski, A., & Hilton, J. L. (2001). Attitudes standing patterns of attitude change: When and the Implicit Association Test. Journal of implicit measures show change, but explicit Personality and Social Psychology, 81, 774– measures do not. Journal of Experimental 788. Social Psychology, 44, 1355–1361. Lee, H., Heller, A. S., van Reekum, C. M., Nel- Gawronski, B., & Strack, F. (2004). On the son, B., & Davidson, R. J. (2012). Amygdala–­ propositional nature of cognitive consistency: prefrontal coupling underlies individual differ- Dissonance changes explicit, but not implicit ences in emotion regulation. NeuroImage, 62, attitudes. Journal of Experimental Social Psy- 1575–1581. chology, 40, 535–542. Lieberman, M. D. (2007). Social cognitive neu- Gilbert, D. T. (1991). How mental systems roscience: A review of core processes. Annual believe. American Psychologist, 46, 107–119. Review of Psychology, 58, 259–289. Greenwald, A. G., & Banaji, M. R. (1995). Lowery, B. S., Hardin, C. D., & Sinclair, S. Implicit social cognition: Attitudes, self-­ (2001). Social influence effects on automatic esteem, and stereotypes. Psychological racial prejudice. Journal of Personality and Review, 102, 4–27. Social Psychology, 81, 842–855. Greenwald, A. G., McGhee, D. E., & Schwartz, Mauss, I. B., Evers, C., Wilhelm, F. H., & Gross, J. L. (1998). Measuring individual differences J. J. (2006). How to bite your tongue without in implicit cognition: The Implicit Association blowing your top: Implicit evaluation of emo- Test. Journal of Personality and Social Psy- tion regulation predicts affective responding chology, 74, 1464–1480. to anger provocation. Personality and Social Gross, J. J. (1998). The emerging field of emotion Psychology Bulletin, 32, 589–602. regulation: An integrative review. Review of Mayo, R., Schul, Y., & Burnstein, E. (2004). “I General Psychology, 2, 271–299. am not guilty” vs. “I am innocent”: Successful Gross, J. J., & Barrett, L. F. (2011). Emotion negation may depend on the schema used for generation and emotion regulation: One or its encoding. Journal of Experimental Social two depends on your point of view. Emotion Psychology, 40, 433–449. Review, 3, 8–16. McConnell, A. R., Dunn, E. W., Austin, S. N., & Herbert, C., Deutsch, R., Sütterlin, S., Kübler, Rawn, C. D. (2011). Blind spots in the search A., & Pauli, P. (2011). Negation as a means for happiness: Implicit attitudes and nonverbal for emotion regulation?: Startle reflex modu- leakage predict affective forecasting errors. lation during processing of negated emotional words. Cognitive, Affective, and Behavioral Journal of Experimental Social Psychology, Neuroscience, 11, 199–206. 47, 628–634. Hofmann, W., Gawronski, B., Gschwendner, T., McConnell, A. R., & Rydell, R. J. (in press). The Le, H., & Schmitt, M. (2005). A meta-a­ nalysis systems of evaluation model: A dual-­systems on the correlation between the Implicit Asso- approach to attitudes. In J. Sherman, B. Gaw- ciation Test and explicit self-r­eport measures. ronski, & Y. Trope (Eds.), Dual process theories of the social mind. New York: Guilford Press. Personality and Social Psychology Bulletin, Moskowitz, G. B., Gollwitzer, P. M., Wasel, W., 31, 1369–1385. & Schaal, B. (1999). Preconscious control of Hovland, C. I., Janis, I. L., & Kelley, H. H. stereotype activation through chronic egali- (1953). Communication and persuasion: Psy- tarian goals. Journal of Personality and Social chological studies of opinion change. New Psychology, 77, 167–184. Haven, CT: Yale University Press. Olson, M. A., & Fazio, R. H. (2009). Implicit Jones, C. R., & Fazio, R. H. (2010). Person cat- and explicit measures of attitudes: The per- egorization and automatic racial stereotyping spective of the MODE model. In R. E. Petty,

266 SOCIAL ASPECTS R. H. Fazio, & P. Briñol (Eds.), Insights from and increased information processing. Jour- the new implicit measures (pp. 19–63). New nal of Experimental Social Psychology, 44, York: Psychology Press. 1526–1532. Olson, M. A., Fazio, R. H., & Hermann, A. D. Rydell, R. J., McConnell, A. R., Mackie, D. M., (2007). Reporting tendencies underlie discrep- & Strain, L. M. (2006). Of two minds: Form- ancies between implicit and explicit measures ing and changing valence-i­nconsistent implicit of self-e­ steem. Psychological Science, 18, 287– and explicit attitudes. Psychological Science, 291. 17, 954–959. Payne, B. K., Burkley, M. A., & Stokes, M. B. Sassenberg, K., & Moskowitz, G. B. (2005). (2008). Why do implicit and explicit attitude Don’t stereotype, think different!: Overcom- tests diverge?: The role of structural fit. Jour- ing automatic stereotype activation by mind- nal of Personality and Social Psychology, 94, set priming. Journal of Experimental Social 16–31. Psychology, 41, 506–514. Petty, R. E. (2006). A metacognitive model of Scott, W. A. (1962). Cognitive complexity and attitudes. Journal of Consumer Research, 33, cognitive flexibility. Sociometry, 25, 405–414. 22–24. Schwarz, N., & Bohner, G. (2001). The construc- Petty, R. E., & Briñol, P. (2006). A metacognitive tion of attitudes. In A. Tesser & N. Schwarz approach to “implicit” and “explicit” evalua- (Eds.), Blackwell handbook of social psychol- tions: Comment on Gawronski and Boden- ogy: Intraindividual processes (pp. 436–457). hausen (2006). Psychological Bulletin, 132, Oxford, UK: Blackwell. 740 –744. Sloman, S. A. (1996). The empirical case for two Petty, R. E., Briñol, P., & DeMarree, K. G. systems of reasoning. Psychological Bulletin, (2007). The Meta-C­ ognitive Model (MCM) of 119, 3–22. attitudes: Implications for attitude measure- Smith, E. R., & DeCoster, J. (2000). Dual pro- ment, change, and strength. Social Cognition, cess models in social and cognitive psychol- 25, 657–686. ogy: Conceptual integration and links to Petty, R. E., Tormala, Z. L., Briñol, P., & Jarvis, underlying memory systems. Personality and W. B. G. (2006). Implicit ambivalence from Social Psychology Review, 4, 108–131. attitude change: An exploration of the PAST Spivey, M. J. (2007). The continuity of mind. model. Journal of Personality and Social Psy- Oxford, UK: Oxford University Press. chology, 90, 21–41. Strack, F., & Deutsch, R. (2004). Reflective and Roskos-­Ewoldsen, D. R., & Fazio, R. H. (1992). impulsive determinants of social behavior. On the orienting value of attitudes: Attitude Personality and Social Psychology Review, 8, accessibility as a determinant of an object’s 220 –247. attraction of visual attention. Journal of Per- Todd, R. M., Cunningham, W. A., Anderson, sonality and Social Psychology, 63, 198–211. A. K., & Thompson, E. (2012). Affect-b­ iased Russell, J. A. (2003). Core affect and the psycho- attention as emotion regulation. Trends in logical construction of emotion. Psychological Cognitive Sciences, 16, 365–372. Review, 110, 145–172. Watson, D., & Tellegen, A. (1985). Toward a Rydell, R. J., & Durso, G. R. O. (2012). Can I consensual structure of mood. Psychological borrow a feeling?: Spillover of negative arousal Bulletin, 98, 219–235. from inconsistent information during attitude Wilson, T. D., Lindsey, S., & Schooler, T. Y. formation diminishes perceptions of well- (2000). A model of dual attitudes. Psychologi- being. Journal of Experimental Social Psy- cal Review, 107, 101–126. chology, 48, 575–578. Wojnowicz, M. T., Ferguson, M. J., Dale, R., & Rydell, R. J., & McConnell, A. R. (2006). Under- Spivey, M. J. (2009). The self-o­ rganization of standing implicit and explicit attitude change: explicit attitudes. Psychological Science, 20, A systems of reasoning analysis. Journal of 1428–1435. Personality and Social Psychology, 91, 995– Zanna, M. P., & Rempel, J. K. (1988). Attitudes: 1008. A new look at an old concept. In D. Bar-Tal Rydell, R. J., McConnell, A. R., & Mackie, D. & A. Kruglanski (Eds.), The social psychol- (2008). Consequences of discrepant explicit ogy of knowledge (pp. 315–334). New York: and implicit attitudes: Cognitive dissonance Cambridge University Press.

Chapter 17 Emotion Regulation in Couples Robert W. Levenson Claudia M. Haase Lian Bloch Sarah R. Holley Benjamin H. Seider Whereas research on emotion regulation in 2010). These couples almost always have individuals has been extremely useful, there problems with emotion regulation. are numerous advantages to studying emo- tion regulation in couples. Couples afford The Social Nature high ecological validity, provide opportuni- of Emotion Regulation ties to view the rich panoply of emotion reg- Research on emotion regulation in couples ulatory strategies, and are ideal for study- should be booming. After all, most human ing the dynamics of emotion regulation as emotions occur in decidedly social situations partners engage in a rich choreography of (Campos, Walle, Dahl, & Main, 2011). emotional expression and regulation that Other people are deeply entwined in the unfolds in complex ways over time. fabric of our emotional lives. They are the proximal stimulus for most of our emotions In this chapter, we consider several and the recipients of most of the emotions important aspects of emotion regulation in we express. When others respond to our couples, discussing its social nature, defin- emotions with their own emotional reac- ing qualities, consequences, development, tions, they provide the fuel needed to sustain and assessment. We end by considering the chains of exchanged emotions that are so the future of this research area, including emblematic of our social lives. unmet needs, unfilled gaps, and unanswered research questions. Our aim in this review With so much of human emotion being of the relevant literatures is not intended to socially situated, it is reasonable to expect be exhaustive but rather to highlight key that most emotion regulation would be simi- studies that illustrate important issues. We larly social. In fact, it has been observed that also draw anecdotally from our experiences up to 98% of emotion regulation episodes with couples who seek therapy for troubled relationships (Levenson, Cowan, & Cowan, 267

268 SOCIAL ASPECTS may take place in social contexts (Gross, individuals. For example, according to the Richards, & John, 2006). Viewed from a 2009 census, 96% of all Americans over age functionalist perspective, emotion regula- 65 have been married at least once. tion is a critical element for promoting social cohesion. If we were to discharge our emo- Defining Qualities tions upon others, full-bore, undiluted, and There are many ways to define emotion reg- absent the moderating influence of emotion ulation. Because most theory and research regulation, the resultant affective tsunami in the field of emotion regulation have would have dire consequences for us and for focused on individuals, popular definitions our social groupings. of emotion regulation reflect this bias. For example, Gross (1998b, p. 275) defines emo- Given its profoundly social nature, we tion regulation as “the processes by which might expect that studies of emotion regula- individuals [emphasis added] influence tion would almost always be conducted in which emotions they have, when they have social situations. However, this could not them, and how they experience and express be further from the truth. In a recent review these emotions.” Of course, a definition of of studies since 2001 (Campos et al., 2011), this sort could be altered so that “individu- less than 12% of the studies assessed emo- als” becomes “individuals and couples.” tion regulation in the presence of another But does this cover all bases? The answer to person (and this is an optimistic estimate this question depends on whether we think that includes studies using both imagined that emotion regulation in couples can be and real others). Clearly, studies of emotion fully captured by summing the regulatory regulation in individuals have advantages activities of the two individuals involved, or over those conducted with couples. They are whether there are emergent qualities of emo- easier to administer and more amenable to tion regulation that are found only in the tight experimental control. However, they couples context. Clearly there are aspects are not optimal for studying the dynamic, of emotion regulation that are common to interpersonal aspects of emotion regulation. both individuals and couples. For example, Although most theorists would agree that emotion regulation in both can be explicit emotion regulation is profoundly social in (i.e., effortful) or implicit (i.e., automatic) nature and is a fundamental area of concern (Gyurak, Gross, & Etkin, 2011), and suc- in close relationships, the literature on emo- cessful or unsuccessful (Gross & Levenson, tion regulation in couples is still surprisingly 1993). And in both contexts it can be diffi- immature, with many gaps and unanswered cult to determine where emotional reactivity questions. ends and emotion regulation begins (Gross, Sheppes, & Urry, 2011). Nonetheless, as we Before moving forward with our consid- hope the following sections illustrate, emo- eration of emotion regulation in couples in tion regulation in couples has a number of intimate relationships we should note that characteristics that are quite different from they are only one of many kinds of dyads those found in individuals. for whom emotion regulation is critical (e.g., new acquaintances, friends, enemies). Dynamic and Iterative Moreover, emotion regulation in social con- Consider two emotion regulatory scenarios. texts scales up from dyads through families, In the first, a recent PhD recipient is prepar- groups, communities, and nations. Emotion ing to give an important job talk and is in regulation in these larger social groupings is the throes of a bout of stage fright. Fear is fascinating and instructive (e.g., communi- the dominant emotion welling up. He is con- ties and nations coping with fear, anger, and cerned that the intensity of his fear, should grief following collective losses, seen quite it reach sufficiently high levels, will compro- famously in the U.S. and world responses to mise the quality of the talk. The standard-­ the September 11, 2001, terrorist attacks). issue emotion regulatory toolbox offers a So, why focus on couples in intimate rela- number of strategies, including altering the tionships? Dyads are the smallest social unit, thus providing an excellent starting point for building a science of socially embedded emotion regulation. Intimate relationships assume a central role in the lives of most

Emotion Regulation in Couples 269 context, reappraising the situation, or will- reducing emotional responses. In the canon- fully controlling aspects of the emotional ical examples, individuals seek to reduce response. The speaker chooses one (or their negative emotions to avoid harmful more) of these strategies, applies them in the consequences for self and others. Consis- moment, successfully down-r­egulates the tent with this, our own early experimental fear, gives an excellent talk, and averts the studies of emotion regulation focused exclu- crisis. sively on the consequences of having sub- jects reduce their behavioral responses to Now, consider a second scenario involv- emotion-­inducing films (Gross & Levenson, ing emotion regulation in couples. A mar- 1993). It was only later that we began to con- ried couple is talking about the husband’s sider up-r­ egulation of emotion in these kinds pending surgery, and he is clearly experienc- of studies as well (Kunzmann, Kupperbusch, ing high levels of fear. In that moment he & Levenson, 2005). In recent years, espe- wants calming support from his wife. But cially as researchers have explored the neu- she is reacting with a great deal of sadness, ral bases of emotion regulation using patient talking about her concerns that he might die and neuroimaging models, studies including and that their children would go through both up-­regulation and down-­regulation life without a father. Getting from this start- have become more common (e.g., Gyurak, ing point to a place where both husband Goodkind, Kramer, Miller, & Levenson, and wife are feeling less distressed is not as 2012; Ochsner et al., 2004). However, it is straightforward as the situation in the first still the case that when studying emotion scenario, in large part because there are two regulation in individuals, the primary focus actors involved. Both individuals have to is on reducing emotion. react to their own and their partner’s emo- tional state, the impact of each partner’s In the realm of emotion regulation in regulatory attempts (some well-c­ hosen, oth- couples, this emphasis on down-­regulation ers misguided), and the unfolding sequence is less appropriate. There are many times of action and reaction that will occur over when couples need to amplify the magnitude time as the couple works toward achieving of emotion so that it emerges more clearly a state that is more emotionally optimal for against the backdrop of other aspects of both partners. their interaction. This amplification takes many forms, ranging from the exaggerated The second scenario illustrates some of tonality of emotional speech patterns used the ways that emotion regulation in couples by mothers when they communicate with differs from emotion regulation in indi- and soothe their infants (Fernald, 1991) to viduals. In regulation in couples there are the stylized and exaggerated expressions of always two actors and reactors, each with love and affection that are so important in his or her own emotional motivations, goals, courtship rituals. Interestingly, among those strengths, blind spots, and hot buttons. Cou- who seek couple therapy, it is extremely ples often find themselves in a complex emo- common for one partner (usually the woman tional landscape that changes continuously in heterosexual couples) to desire the other as partners express and regulate their own partner (usually the man in these couples) to emotions, respond to each other’s emotions up-­regulate emotion (i.e., expressing emo- and regulatory attempts, and try to regulate tion more often and more clearly). each other’s emotions. This extremely fluid situation, replete with highly dynamic and Bivalent iterative sequences of emotion, creates an The prototypical examples of emotion reg- extremely challenging, complex landscape ulation in individuals all focus on negative for emotion regulation, one that is quite dif- emotions. This is seen in common parlance ferent from that faced by individuals. (e.g., “Don’t let them see you sweat” [fear], “Don’t let it get to you” [anger], “Grown Bidirectional men don’t cry” [sadness]); in many ado- Although most discussions of emotion reg- lescent rituals in which a calm demeanor ulation in individuals allow for both up-­ is maintained while viewing, touching, or regulation and down-­regulation of emotion, ingesting extremely gross things (disgust); the emphasis most often seems to be on

270 SOCIAL ASPECTS and even in therapeutic contexts (where We expect that the same is true of couples. cathartic release is often sought for pent up Maintaining an optimal level of emotional anger, fear, and sadness). Consistent with arousal for couples, however, requires moni- this emphasis, most laboratory research on toring and regulating the emotional state emotion regulation has focused on control- of both partners. The dynamic and itera- ling negative emotions. Such work is facili- tive nature of emotion exchanges in couples tated by eliciting negative emotions such as means that the level of emotional arousal is disgust readily and powerfully, using static constantly changing. Thus, the maintenance images (Lang, Greenwald, & Bradley, 1988) of an optimal state requires continuous and films (Gross & Levenson, 1995). In addi- monitoring of arousal levels and continuous tion, although studies comparing the regula- adjustment of regulatory efforts. This situ- tion of negative and positive emotions have ation has caused us on multiple occasions been rare, these comparisons have often not to observe that a good marriage requires a produced dramatic differences (e.g., Gross good thermostat (a role most often assumed & Levenson, 1997). by wives in heterosexual couples; Gottman & Levenson, 1988). In couples, however, the regulation of positive emotion is at least as important Further complicating matters, in cou- as the regulation of negative emotion. Up-­ ples, one partner’s regulatory efforts often regulating positive emotion is critical for become potent emotional stimuli for the building and maintaining intimate relation- other partner. Consider this example: A ships throughout the lifespan. This is seen husband has been trying to soothe his wife’s in parent–­infant and parent–­child relation- anger over a canceled vacation, hoping that ships, and continues with childhood friend- when she is calmed down they will be able ships, early romantic relationships, mate to discuss alternative plans. At this point in selection, and long-term committed rela- the interaction, the wife is overly aroused tionships (Carstensen, Graff, Levenson, & and the husband is relatively calm. The wife Gottman, 1996). Similarly, down-r­ egulating says, “I hate it when you try to manage me. positive emotion also plays an important role It’s insulting and belittling. You’re just so for couples. For example, failure to down-­ incredibly selfish; if you really loved me, you regulate amusement in response to a part- wouldn’t do this to me.” Hearing this, the ner’s failures and insecurities can be critical. husband feels unjustly judged and injured. In these contexts, laughing, aggressive teas- His anger starts welling up and he becomes ing, and unrelenting, unsupportive humor quite defensive. Suddenly, they are both can be experienced as cruel and demeaning, overly aroused and find themselves casting thus serving to undermine relationship qual- hurtful aspersions about each other’s char- ity (see Martin, 2007). acter defects. Any possibility of having a constructive discussion about vacation plans Coregulatory will soon be placed on indefinite hold. Most of us find managing our own emotions sufficiently challenging to occupy signifi- Coregulation in couples would be chal- cant segments of our waking (and sleeping) lenging enough if both partners always had hours. This challenge increases dramatically the same regulatory goals (e.g., both want- when we take on the additional responsibil- ing to feel less aroused and more calm, or ity of attempting to manage the emotions both wanting to intensify feelings of pas- of another person. Whether the interaction sionate love). However, the emotional partner is an infant, a friend, a romantic love stars are not always so well-­aligned. The interest, or a partner in a long-term relation- “demand–­withdraw” pattern, commonly ship, moving the focus of emotion regula- found in both opposite sex (Christensen, tion from “my emotions” to “your and our 1987) and same-sex (Holley, Sturm, & emotions” takes us into new and complex Levenson, 2010) couples, provides a good realms. A recurrent theme in studies of opti- example. A couple wants to talk effectively mal performance is that individuals function about a significant relationship issue. How- optimally when they are somewhat, but not ever, one partner (typically the one who overly, aroused (Yerkes & Dodson, 1908). wants change) becomes quite aroused and engages energetically in complaining and criticizing the other partner. The criticized

Emotion Regulation in Couples 271 partner (typically the one who wants to with marriage and other committed rela- maintain the status quo) increasingly tunes tionships, has produced a number of find- out and withdraws emotionally from the ings that suggest greater ability to regulate interaction, thus achieving some level of emotion in couples is associated with posi- calm. For this couple, the coregulatory chal- tive outcomes. For example, questionnaire lenge is to calm the overaroused, demanding studies have found that couples who report partner and at the same time increase the less frequent use of “control or contain- emotional involvement of the underaroused, ment” (which is similar to suppression) of withdrawing partner. However, this must negative emotion have higher marital satis- be done without causing either partner to faction (Feeney, 1999). In the literature on overshoot the desired emotional endpoint intimate partner violence, inability to regu- (no easy feat). Coregulation of emotion in late negative emotion has been associated couples regularly introduces these kinds of with increased likelihood of partner abuse complexities, which are simply not found in (McNulty & Hellmuth, 2008). individual emotion regulation. A paradigm for studying couples’ interac- Consequences tion developed by Levenson and Gottman It is easy to make a case for the importance of (1983) has been endorsed as an exemplar emotion regulation in the lives of individuals. for how emotion regulation can be studied Greater use of specific emotion regulation in social contexts (Campos et al., 2011). In strategies, as measured by self- and other-­ this paradigm, couples (usually married het- reports, has been found to predict higher erosexual couples, but also same-sex and levels of well-being, mental health, physical dating couples) come to the laboratory and health, relationship quality, and social func- engage in a series of unrehearsed 15-minute tioning, and lower levels of problem behav- conversations on relationship topics (e.g., ior (Aldao, Nolen-H­ oeksema, & Schweizer, events of the day, a problem area, a pleasant 2010; Caspi, Henry, McGee, Moffitt, & topic). During these conversations, behavior Silva, 1995; Gross & John, 2003; John & is videotaped for subsequent coding of emo- Gross, 2004; Lopes, Salovey, Côté, Beers, tional behavior by trained observers, and & Petty, 2005; Nelis et al., 2011). Although in both partners a number of physiological studies of these associations using labora- measures relevant to emotional responding tory assessments of emotion regulation (as are monitored continuously. Later, partners opposed to self- and other-­reports) are still view the videotapes of their conversations rare, we recently found that greater ability and use a rating dial to provide continu- to down-­regulate and up-­regulate emotional ous ratings of the valence of their emotional response (as assessed using well-­established experience during the interactions (Gott- laboratory procedures) was associated with man & Levenson, 1985). These streams of greater well-being and higher income (Côté, continuous multimethod data (self-r­eport, Gyurak, & Levenson, 2010). If emotion behavior, physiology) can be used to derive regulation is broadened to include delay of measures of emotion reactivity and emotion gratification, another laboratory-­based par- regulation in the individual partners and in adigm, research strongly indicates that high the dyad. levels of this ability early in preschool years are associated with a host of positive out- A number of findings from these stud- comes, including greater cognitive and aca- ies seem highly relevant when considering demic competence and ability to cope with the consequences of emotion regulation frustration later in life (Mischel et al., 2011). for couples. In this regard, the discussions about marital problems (which can occa- Because of a relative dearth of studies sion intense negative emotions and heroic that have directly measured emotion regu- efforts at emotion regulation) have been lation in couples and its consequences, the particularly informative, with links found case for the importance of emotion regula- between measures of emotion regulation (in tion in couples must be based on collateral subjective experience, behavior, and physiol- literatures. One such literature, concerned ogy) and important consequences in several domains. In terms of subjective emotional experi- ence, low levels of both negative emotion

272 SOCIAL ASPECTS and negative emotion reciprocity (i.e., nega- coldness and lack of empathy, and absence tive emotional experience by one partner of joy) or both. Historically, couples thera- followed by negative emotional experience pies have focused more on nonemotional by the other partner) have been associated aspects of these problems (e.g., communi- with higher levels of marital satisfaction cation deficits, individual psychopathology, both concurrently and over time (Leven- attachment histories, poor family-­of-­origin son & Gottman, 1983, 1985). In terms of relationship models), but emotion and emo- emotional behavior, husbands’ inability to tion regulation are increasingly becoming deescalate negative emotion during marital important foci in many forms of couple ther- conflict predicted less marital stability over apy (Gottman & Gottman, 2008; Johnson, time (Gottman, Coan, Carrere, & Swanson, 1996; Wile, 2002). 1998). In a similar vein, “regulated” cou- ples, operationalized as those who produced Development an increasingly high ratio of positive to neg- Viewed from a developmental perspective, ative emotional behaviors over the course several key dyadic relationships assume of a 15-minute conflictive interaction, had prominence at different stages of the lifes- higher levels of marital satisfaction, lower pan and serve as crucibles for the emergence risk for marital dissolution, and better health and refinement of regulatory skills. Here we measured over a 4-year period (Gottman & focus on three of these dyads: parents and Levenson, 1992). Finally, less escalation of infants, early romantic relationships, and negative emotional behavior was associated late-life couples. with higher marital satisfaction in couples in long-term marriages (Carstensen, Gottman, Parent–Infant Dyads & Levenson, 1995). Parent–­infant dyads, and mother–­infant dyads in particular, invest significant efforts The continuous measures of peripheral in emotion regulation. The initial focus is physiological activity have been particularly on reducing negative emotion (managing interesting, in part because autonomic ner- the infant’s distress), but this quickly engen- vous system responses are very difficult to ders efforts to increase positive emotion control voluntarily (e.g., Levenson, 1976) (engaging in activities that amuse, distract, and because they can provide a ready metric and calm). In this stage of life, infants can of changing levels of arousal in the couple. become overwhelmed by their negative emo- Using these measures, low levels of both tions and lack the skills to bring them under physiological arousal and physiological control themselves. Thus, infants rely on linkage (synchrony between partners’ physi- their caregivers to regulate their emotions ology) were associated with higher marital (Thompson, 1991). The ontological origins satisfaction (Levenson & Gottman, 1983, of social emotion regulation clearly reside in 1985). these dyads. If all goes well, coregulation of emotion in the parent–i­nfant dyad will lead Couple Therapy to infants beginning to develop the ability to Not surprisingly, given its critical role in regulate their own emotions. couple relationships and profound down- stream consequences, issues with emotion Attachment theory (Bowlby, 1988) pro- regulation frequently assume center stage vides the most influential account of the when couples seek treatment for relationship transition from coregulation to individual problems. Although the specifics differ from regulation in infancy, providing elegant couple to couple, distressed couples almost descriptions about the kinds of parent–­ always struggle with either down-­regulating infant relationships that are likely to result negative emotion (e.g., issues involving jeal- in good versus poor emotion regulation in ousy and disagreements over things such the infant. Attachment theory was given an as household duties, relatives, child rear- enormous empirical boost with the develop- ing, and finances) or up-r­egulating positive ment of observational methods for quantify- emotions (e.g., issues involving poor and ing mother–­infant attachment. For example, infrequent communication, not doing things together, loss of sexual interest and intimacy,

Emotion Regulation in Couples 273 the “Strange Situation” paradigm (Ain- Early Romantic Dyads sworth, Blehar, Waters, & Wall, 1978) uses Selecting a partner and building a roman- close observation of episodes of mother–­ tic bond are critical developmental tasks. infant separation and reunion to classify Although often viewed through a lens that attachment styles in ways that have profound emphasizes mate selection and family build- implications for emotion regulation initially ing (Havighurst, 1976), these relationships in the dyad and ultimately in the infant (e.g., are also important vehicles for develop- securely attached infants develop greater ing emotion regulatory skills. In contrast ability to regulate their own emotions than to parent–­infant dyads, where the primary do insecurely attached infants). focus is on mastering down-r­egulation of negative emotions, in early romantic rela- The Strange Situation focuses on the tionships the primary focus is clearly on down-­regulation of negative emotions (pri- up-r­egulating positive emotions. The emo- marily fear and sadness), but the mother–­ tions that are typically targets for this up-­ infant dyad also is involved in a great deal of regulation include passionate love, affection, coregulation of positive emotion. The social joy, excitement, and enthusiasm (Gable, nature of positive emotion regulation in Gonzaga, & Strachman, 2006). infancy is seen vividly in studies of the syn- chrony and reciprocity of positive emotion Several lines of research highlight the (Cole, Teti, & Zahn-­Waxler, 2003; Tron- importance of positive emotions in early ick, 1989). In addition, parent–­infant dyads romantic relationships. Romantic love is a often engage in elaborate behavioral rituals phenomenon found across many cultures designed to increase positive emotions in the (Jankowiak & Fischer, 1998) and has been infant (e.g., tickling and peekaboo rituals). termed a “mammalian system for mate For parents, the infant’s smile is a highly choice” (Fisher, Aron, & Brown, 2006). prized reinforcer; parents will go to great It is associated with a complex physiologi- lengths to evoke smiles in their infants. cal, psychological, and behavioral profile that includes feelings of euphoria, focused We include discussion of these early attention, and obsessive thinking about a parent–­infant dyads because we believe they specific individual; craving for emotional have important implications for emotion connection with the other person; expanded regulation and other aspects of later inti- sense of self; and greatly increased energy mate relationships. However, one enduring (A. Aron et al., 2005; E. N. Aron & Aron, question about attachment styles (and asso- 1996). There is good evidence linking the ciated emotion regulatory abilities) is how early stages of intense passionate love with parent–­infant interactions are related to subcortical reward and goal centers in the attachment styles that characterize intimate brain that are highly responsive to dopa- relationships later in life (Mikulincer & mine (A. Aron et al., 2005; Fisher, Aron, Shaver, 2007). In this regard, some theorists & Brown, 2006), which may help explain have emphasized discontinuity (e.g., Kagan, the almost addictive quality of passionate 1984), while others have leaned more toward love, along with its attendant cravings (e.g., continuity (e.g., Bowlby, 1988). Conducting intensely missing the partner when absent) empirical research on these issues is quite and powerful withdrawal reactions (e.g., the challenging. One approach has been to uti- pain of lost love). From an emotion regula- lize chain mediation models, in which expe- tory point of view, these high-­intensity posi- riences during one life stage predict expe- tive emotional states are highly desirable, riences in the next, which in turn predict and, not surprisingly, lovers engage in quite experiences in the next (Sroufe, Coffino, & elaborate strategies to up-r­egulate their Carlson, 2010). In one such study, which positive feelings to extremely high levels bridged infancy, adolescence, and adult rela- of intensity. There is little doubt that these tionships, secure attachment in infancy was intense feelings contribute significantly to found to predict more secure relationships mate selection and reproduction. In many with close friends in adolescence, which in ways, they are the perfect fuel for launch- turn predicted more positive daily emotional ing romantic dyads along the path to family experiences in adult romantic relationships formation. and less negative affect in conflict resolution (Simpson, Collins, Tran, & Haydon, 2007).

274 SOCIAL ASPECTS Clearly, down-r­egulating negative emo- and loss (Mather, 2012). In survey studies, older individuals report believing that they tion is an important item on the dyadic improve in this ability (Gross et al., 1997). agenda at all stages of development. This Laboratory studies (e.g., Shiota & Leven- often takes the form of managing jeal- son, 2009) paint a more complex picture, ousy (Shaver & Mikulincer, 2007) in early with some regulatory strategies remaining romantic relationships, in which real and stable with age (i.e., suppressing visible signs imagined infidelities are a source of power- of emotional response), others declining ful negative emotions (fear, sadness, anger) with age (i.e., using detached appraisals to that must be controlled if the relationship down-r­egulate emotion), and still others in is to survive. Emotion regulation continues fact improving (i.e., using positive apprais- to play a critical role as romantic relations als to down-r­egulate emotion). In late life, grow and develop. Looming large is the tran- up-r­egulating particular kinds of positive sition to parenthood, which is highly chal- emotion is very important. For example, lenging for most couples (Cowan & Cowan, reminiscing with others about past accom- 1992; Doss, Rhoades, Stanley, & Markman, plishments can be particularly rewarding 2009). Successful navigation of this transi- (Erikson, 1982), and companionate love, tion requires couples to deploy the full range which is characterized by low levels of pas- of individual, dyadic, and, ultimately, triadic sion but high levels of intimacy and com- emotion regulatory skills. mitment (Sternberg, 1986), often assumes the position of primary importance that Late‑Life Dyads was occupied earlier by romantic love. Also In late life, as couples move beyond the important is the ability to down-r­egulate prime reproductive period, life challenges particular negative emotions such as sad- and life goals change, and emotion regula- ness (in response to interpersonal losses) and tory needs change accordingly. During this embarrassment (in response to losses in cog- developmental period, dealing with losses nitive and physical abilities). Finally, com- and finding meaning in life become particu- plex emotions such as poignancy, become larly salient (Erikson, Erikson, & Kivnick, increasingly prevalent in late life (Ersner-­ 1986). As individuals get older, they expe- Hershfield, Mikels, Sullivan, & Carstensen, rience functional losses in domains such 2008), leading to time spent reliving and as cognition (Salthouse, 2004), physical savoring memories that have both positive abilities, and health (albeit with consider- and negative emotional qualities (e.g., chil- able individual differences, Rowe & Kahn, dren marrying and leaving home). 1997). They also experience losses in their social networks as retirement from the Importantly, the coregulation of emo- workforce limits daily social contacts and tion also becomes increasingly important friends are lost due to relocation, illness, as older adults spend more time with their and death (Charles & Carstensen, 2007; spouses as opposed to friends and acquain- Wrzus, Hänel, Wagner, & Neyer, 2013). tances (Charles & Carstensen, 2007). For Lifespan developmental theory emphasizes those without close friends, down-r­ egulating the importance of finding new sources of negative emotions (e.g., sadness, fear) asso- meaning in late life, because earlier sources ciated with loneliness becomes critically of meaning (e.g., partner selection, family important. The stakes may be especially building, career building) are no longer as high for failures of emotion regulation in relevant. These can include developing qual- late life; the negative consequences of lone- ities of generativity (caring for future genera- liness, for example, on health, have been tions) and integrity (acceptance of one’s life) widely documented (Hawkley & Cacioppo, (Erikson, 1950), and investing more deeply 2010). in close social relationships (Carstensen, Isaacowitz, & Charles, 1999). Earlier we presented evidence that cou- ples’ ability to lower levels of physiologi- In late-life couples, emotion regulation cal arousal is an important predictor of becomes very important and may contrib- relationship quality and stability over time ute to older adults’ relatively preserved lev- (Gottman & Levenson, 1992; Levenson & els of well-being even in the face of decline Gottman, 1985). These kinds of calming effects can be produced by touch (Coan,

Emotion Regulation in Couples 275 Schaefer, & Davidson, 2006) and also by clearly useful, are a far cry from the kinds of positive emotions, which can reduce levels intimate dyads in which emotion regulation of autonomic nervous system arousal pro- emerges and is refined during development duced by negative emotions in individu- (see earlier discussion). als (Fredrickson & Levenson, 1998) and in couples (Yuan, McCarthy, Holley, & Observational studies of mother–i­nfant, Levenson, 2010). These soothing effects of mother–­toddler, and mother–­preschooler positive emotions may contribute to positive dyads (Cole et al., 2003; Denham, 1993; emotions becoming increasingly important, Dumas, LaFreniere, & Serketich, 1995; desired, and salient in late life (Carstensen Tronick, 1989) provide one bright spot in et al., 1999). this otherwise sparsely populated land- scape. Although typically not designed to Assessment study emotion regulation per se, they do Emotion regulation in couples can be mea- offer some valuable insights as to how emo- sured using self-­report measures that assess tion regulatory processes function in these beliefs (one’s own or those of others who early-life dyads. know us) about emotion regulation. Alter- natively, emotion regulation in couples can Self‑Report Measures be measured using performance measures In the emotion regulation domain, most based on the observation of actual emotion self-r­eport measures focus on regulation in regulation. In both self-r­eport and perfor- the individual, not in the couple. For exam- mance measures, the focus can be on regula- ple, the Emotion Regulation Questionnaire tory abilities (i.e., what the person or couple (Gross & John, 2003) assesses two styles of is capable of doing) or regulatory practices emotion regulation using a 10-item scale in (i.e., what the person or couple typically which six items assess the dispositional ten- does). As noted earlier, there are aspects (e.g., dency to use cognitive reappraisal strategies coregulation) and qualities (e.g., dynamic) of and four items assess the dispositional ten- emotion regulation that are more prominent dency to use suppression strategies. Another in couples than in individuals. Thus, the individual-f­ocused inventory, the Difficul- measures used for assessing emotion regula- ties in Emotion Regulation Scale (Gratz & tion in couples should be designed to capture Roemer, 2004), comprises 36 items that these qualities, in addition to qualities that assess six dimensions of emotion regulation: are also prominent in emotion regulation in (1) lack of awareness of emotional responses, individuals. (2) lack of clarity of emotional responses, (3) nonacceptance of emotional responses, (4) Unfortunately, the state of the art in mea- limited access to emotion regulation strat- suring emotion regulation in couples is not egies, (5) difficulties controlling impulses as advanced as we would wish. As we discuss when experiencing negative emotions, and below, most existing self-­report measures of (6) difficulties engaging in goal-d­irected emotion regulation clearly focus on indi- behavior when experiencing negative emo- vidual regulation. The self-­report measures tions. that do have items relevant for assessing emotion regulation in couples were almost It is certainly possible to alter an individ- all designed for other purposes (e.g., mea- ual self-r­eport measure of emotion regula- suring relationship satisfaction). Progress tion so that items refer to a particular couple in developing methods for assessing actual relationship. However, this would essentially regulatory performance in couples is also constitute a new instrument, and its reliabil- hindered by the lack of studies of emotion ity and validity would need to be established. regulation that include an actual (or even Moreover, questionnaires that were origi- an imagined) interaction partner (Campos nally developed with a focus on the individ- et al., 2011). Moreover, when such stud- ual are unlikely to assess aspects of emotion ies have been conducted, they have often regulation that are particularly relevant to used unacquainted dyads (e.g., Butler et al., couples (e.g., reciprocity of emotion, reac- 2003). These stranger pairings, although tions to each other’s regulatory styles). The literature on close relationships has produced a number of self-­report inventories

276 SOCIAL ASPECTS that were developed to assess relationship capturing enduring dispositional qualities of functioning and satisfaction. A subset of couples’ regulatory styles but are not well-­ these inventories focuses on couples’ conflict suited for capturing the dynamics of emo- resolution and communication skills, and tion regulation. Emotion self-­reports can be includes items that are relevant to assessing obtained in ways that yield more dynamic couples’ emotion regulation. For example, in information using affect rating dial (Ruef the 78-item Revised Conflict Tactics Scales & Levenson, 2007) and experience sam- (Straus, Hamby, Boney-McCoy, & Sugar- pling (Hektner, Schmidt, & Csikszentmih- man, 1996), respondents are asked about alyi, 2007) methodologies. There have also how they deal with disagreement, with ques- been attempts to use self-­report measures of tions asking how often they “shouted at or emotion regulation in new ways to capture yelled at my partner,” or how often conflict dynamic dyadic processes. For example, escalated into several more violent acts. In in one study (Butner, Diamond, & Hicks, the 109-item Managing Affect and Differ- 2007), couples were asked to provide daily ences Scale (Arellano & Markman, 1995), ratings of positive and negative affect for subscales that are directly relevant to emo- 3 weeks, operationalizing coregulation as tion regulation include Negative Escalation covariation in partners’ daily levels of affect (e.g., “Unable to get out of heated argu- and coupling of the rates of change of part- ments”), Stop Actions (e.g., “When conflicts ners’ affective cycles. In a similar vein, Fer- get out of hand, agree to stop and talk at a rer and Nesselroade (2003) assessed emo- later time”), and Withdrawal (e.g., “When tional experience of partners in one married discussing issues, my partner usually with- dyad, who recorded their experience of 20 draws for fear of conflict”). In a brief, emotions over 182 consecutive days. 10-item screening inventory for relationship discord (Whisman, Snyder, & Beach, 2009), None of these approaches provides an respondents are asked, “Whenever you are off-the-shelf solution for measuring emo- feeling sad, does your partner make you feel tion regulation in couples, but they do pro- loved and happy again?” and “Do minor vide items and scales (mostly focused on disagreements with your partner often end the management of negative emotion) that up in big arguments?” could be useful starting points for build- ing more comprehensive couple assessment There are a few self-­report measures that instruments. Nonetheless, our enthusiasm do focus on individuals’ emotion regulation for self-r­eport measures of emotion regula- vis-à-vis a specific partner (e.g., a roman- tion is tempered by the problems that beset tic partner or a parent). Arguably, the best all self-­report measures (e.g., social desir- known of these are primarily designed to ability and other self-p­resentation biases, assess attachment styles, but they do have vulnerability to wording and context). In items of relevance to emotion regulation. the realm of emotional functioning, these The Experiences in Close Relationships problems are compounded by individuals Scale (Brennan, Clark, & Shaver, 1998) is a often not being very accurate observers and 36-item measure that assesses avoidance and reporters of the nuances of their own emo- anxiety in close relationships, with items tional functioning. Evidence of this comes that focus on the dyadic context (e.g., “I pre- from findings of low correlations between fer not to show a partner how I feel deep self-­report and behavioral measures of emo- down”). The Adult Attachment Interview tional functioning (e.g., low correlations (Main & Goldwyn, 1984) is based on 20 between well-e­stablished self-r­eport mea- questions, some of which focus on impor- sures of empathy and a performance-b­ ased tant aspects of emotion regulation (e.g., measure of empathic accuracy; Levenson experiences with parents involving distress, & Ruef, 1992) and variations among indi- separation, rejection, or loss). It analyzes viduals in the coherence between self-­report, responses to these questions, focusing on not behavioral, and physiological measures of only what is said but also how it is said (e.g., emotion (Mauss, Levenson, McCarter, Wil- the coherence of the narrative). helm, & Gross, 2005; Sze, Gyurak, Yuan, & Levenson, 2010). Whether these kinds of All of the measures reviewed thus far yield problems do in fact beset particular ques- scores that are considered to represent trait- tionnaires designed to measure emotion reg- like qualities. Such approaches are ideal for

Emotion Regulation in Couples 277 ulation in couples, and the ultimate utility of mance measures of emotion regulation are such questionnaires for particular purposes, likely subserved by different neural circuits are best settled on the basis of actual data. (Goodkind, Gyurak, McCarthy, Miller, & Levenson, 2010). For all of these reasons, Performance Measures measures of abilities and performance Whereas self-r­eport measures of emo- should not be used interchangeably. tion regulation assess the rater’s beliefs about emotion regulation in self or others, Measures of regulatory ability that have performance-­based measures assess emo- been widely used in studies of emotion regu- tion regulation in vivo, evaluating emotion lation in individuals instruct participants to regulation as it actually occurs. These are suppress and amplify behavioral responses often multimethod performance measures (e.g., Gross & Levenson, 1993, 1997; Hage- that include assessment of subjective experi- mann, Levenson, & Gross, 2006; Roberts, ence, emotional behavior, and physiological Levenson, & Gross, 2008) and use vari- activation. ous appraisal strategies (e.g., Gross, 1998a; Shiota & Levenson, 2009). However, these As noted earlier, performance measures measures have not been used as much with (and self-r­eport measures) can focus on couples. One exception is a study by Rich- either regulatory abilities or regulatory ards, Butler, and Gross (2003), in which dat- practices. Measures of regulatory abilities ing couples were instructed to reappraise or typically provide participants with spe- suppress their emotions as they discussed a cific instructions to alter some aspect of relationship conflict to determine the effects emotional responding (e.g., “Try not to let on memory for conversation utterances your emotions show”), then assess how well (which were increased by reappraisal and they do (e.g., measuring the amount of vis- decreased by suppression) and emotional ible emotional expression using a sensitive memories (which were increased by suppres- behavioral coding system). Measures of reg- sion). ulatory practices, in contrast, typically place participants in a situation in which emo- Measures of regulatory practices have tion regulation would be expected to occur been utilized in studies of emotion regula- (e.g., discussing an area of conflict with a tion in adult dyads, with emotion regulation relationship partner), then assess how well operationalized in a number of different participants regulate their emotions (e.g., ways, including (1) the amount of negative measuring positive and negative emotional or positive emotional experience (e.g., feel- behaviors and physiological arousal). ing positive regard; Murray, 2005); (2) the amount or ratio of negative and positive Each of these approaches has advantages emotional behavior (Gottman, 1993; Gott- and disadvantages. For example, measures man & Levenson, 1992); (3) autonomic ner- of abilities tell us a lot about what people vous system activation (Levenson & Gott- are capable of doing in the emotion regula- man, 1985); and (4) central nervous system tory domain but not necessarily what they activation (Coan et al., 2006). actually do in their day-to-day lives (e.g., a person might be able to suppress all vis- Measures of regulatory practices have also ible signs of emotion when watching a sad been utilized in mother–­infant dyads, most movie but be very volatile in interactions famously in research using the Strange Situ- with her romantic partner, or vice versa). ation (Ainsworth, Blehar, Waters, & Wall, With measures of practices, the lack of 1978). In this test, children’s behavior is instructions reduces experimental control, measured after separation from the attach- and inferential leaps must be made between ment figure, a prototypical situation for elic- what is measured and its relationship to iting distress and fear in infants. Another emotion regulation (e.g., slowed heart rate measure of emotion regulation in infant–­ might indicate emotional calming, but it caregiver dyads is the Still-Face Paradigm can also have many different psychological (Tronick, Als, Adamson, Wise, & Brazelton, and physiological causes). We should also 1978). Here, the primary caregiver becomes note that recent evidence from neurological unresponsive and maintains a neutral facial patients indicates that ability and perfor- expression. In response, infants typically show increased gaze aversion, less smiling, and heightened negative affect.

278 SOCIAL ASPECTS Many of the aforementioned approaches of couples’ relationships and using couples’ to measuring emotion regulation in couples relationships as a test bed for increasing our result in scores that represent regulation understanding of what emotion regulation averaged over some period of time. However, is, how it operates, and its sources and con- it is also possible for performance-­based sequences. For this research area to move measures to be dynamic, quantifying pat- forward and realize its potential, there are terns of emotional reactivity and regulation several pressing needs: that unfold over time. These more dynamic approaches have been utilized in studies 1.  Needed are sound self-r­eport and per- that track changes in emotional experience formance measures of couples’ emotion reg- (e.g., Levenson & Gottman, 1983), emo- ulation that have been carefully constructed tional behavior (see review in Gottman & and have well-­established psychometric Levenson, 1988), autonomic nervous system qualities of reliability and validity. Mea- physiology (Yuan et al., 2010), hormonal sures are needed that (a) assess both emo- responses (Laurent & Powers, 2007), and tion regulatory abilities (what people can central nervous system activation (Coan et do) and practices (what people do do); (b) al., 2006). Particularly promising are those move beyond a primary focus on the down-­ techniques that use the responses of both regulation of negative emotion to include the members of the dyad to characterize quali- up-r­egulation and down-r­egulation of both ties of emotional coregulation over time positive and negative emotion; (c) assess (Butler, 2011). These include measures of multiple aspects of emotion regulation (sub- (1) emotional reciprocity—the exchange of jective experience, behavior, physiology); emotions between partners in continuous and (d) allow assessment of emotion regula- self-­ratings of emotional experience (Lev- tion in ways that capture its dynamic, itera- enson & Gottman, 1983) or observational tive, co-­regulatory nature. coding of emotional behavior (Carstensen et al., 1995; Gottman et al., 1998; Gottman 2. Needed are the development and & Levenson, 1992; Julien, Brault, Char- refinement of experimental paradigms that trand, & Bégin, 2000; Tronick et al., 1978); are appropriate for studying emotion regu- (2) emotional linkage—the extent to which latory abilities and practices in couples. We physiological, hormonal, or mood responses believe that observational studies of couple of partners become “synchronized” or can and parent–­child interactions provide good be predicted from each other (Levenson & bases for moving forward, but hope that Gottman, 1983; Saxbe & Repetti, 2010); new paradigms will also be developed that or (3) physiological soothing—the transi- stimulate new research in this area. tion from high to low levels of autonomic activation in a couple (Yuan et al., 2010). 3. Relationships between self-­report and Recently, Levenson (2013) described a new performance measures of emotion regula- statistical approach that characterizes cycles tion in couples need to be studied and estab- of emotion (transitions between high and lished. The existing self-­report measures low arousal, and between negative and posi- of emotion regulation in individuals have tive emotion) that incorporates continuous generally not been studied in this way; thus, measurement of behavior, physiology, and there is not a firm basis for assuming equiva- subjective experience. lencies across methods. Agenda for Future Research 4. Issues concerning the distinctions Research on emotion regulation in couples between emotion reactivity and emotion is poised for growth and discovery. Studies regulation need to be addressed in emotion of close social relationships are dramatically regulation in couples, in much the same way increasing in many areas of psychology. as they are being addressed in research on Clearly, we have only scratched the surface individual emotion regulation (e.g., Ochsner in terms of both understanding the role that et al., 2009). These are challenging issues emotion regulation plays in the functioning with deep theoretical and practical impli- cations. They should benefit greatly from expanding knowledge about how emotional functioning is organized in the central and peripheral nervous systems.

Emotion Regulation in Couples 279 5. Research is needed on the anteced- devoted to proffering parenting advice, the ents of emotion regulation in couples. We research needs are similar. expect that additional progress can be made in exploring biological (e.g., genetic, temp­ Although the state of the science in emo- eramental) and psychological (e.g., per- tion regulation in couples is still relatively sonality, attachment history) factors that immature, the potential is clearly enormous. predispose couples to develop particular With the development of new measurement regulatory styles. Of great interest are the tools and sound, ecologically valid experi- ways that couples’ emotion regulatory styles mental paradigms that enable us to study develop over the course of the relationship, regulatory dynamics in interpersonal con- when they are malleable, and when they texts, we expect great progress to be made solidify. in many important areas related to emotion regulation in couples. 6.  More research is needed on the conse- quences of emotion regulation, with special References attention given to mapping different kinds of emotion regulation in relation to outcomes Ainsworth, M. D. S., Blehar, M. C., Waters, E., in multiple domains (including physical and & Wall, S. (1978). Patterns of attachment: A mental heath, well-being, and relationship psychological study of the Strange Situation. quality). This kind of research would benefit Oxford, UK: Erlbaum. greatly from longitudinal designs; we expect that many effects of particular regulatory Aldao, A., Nolen-­Hoeksema, S., & Schweizer, S. styles only emerge over fairly long periods (2010). Emotion-r­egulation strategies across of time. psychopathology: A meta-­analytic review. Clinical Psychology Review, 30, 217–237. 7. Research is needed on the similarities and differences that emerge when socially Arellano, C. M., & Markman, H. J. (1995). situated emotion regulation is scaled up The Managing Affect and Differences Scale from dyads to larger groups (e.g., commu- (MADS): A self-r­ eport measure assessing con- nity responses to traumatic events). We need flict management in couples. Journal of Fam- more comparisons of the nature of emotion ily Psychology, 9, 319–334. regulation in different kinds of dyads (e.g., romantic, friendship, coworker dyads). We Aron, A., Fisher, H., Mashek, D. J., Strong, G., also need more research on how communi- Li, H., & Brown, L. L. (2005). Reward, moti- ties deal with tragic and triumphant events, vation, and emotion systems associated with with a particular focus on the emotion regu- early-stage intense romantic love. Journal of latory challenges that must be faced in their Neurophysiology, 94, 327–337. aftermath. Aron, E. N., & Aron, A. (1996). Love and the 8.  Research is needed on ways to improve expansion of the self: The state of the model. emotion regulation abilities in individuals, Personal Relationships, 3, 45–58. dyads, and groups. In the realm of adult rela- tionships, there are a number of therapeutic Bowlby, J. (1988). A secure base: Parent–child approaches that afford particular attention attachment and healthy human development. to emotional functioning (Gottman & Gott- New York: Basic Books. man, 2008; Johnson, 1996; Wile, 2002). Unfortunately, research on the efficacy and Brennan, K. A., Clark, C. L., & Shaver, P. R. effectiveness of these treatments tends to (1998). Self-r­eport measurement of adult compare outcomes for one extremely com- attachment: An integrative overview. In J. A. plex, multifaceted treatment package to a Simpson & W. S. Rholes (Eds.), Attachment non-­treatment condition (or sometimes to theory and close relationships (pp. 46–76). another highly complex treatment pack- New York: Guilford Press. age). More research is needed using designs that enable honing in on the “active ingredi- Butler, E. A. (2011). Temporal interpersonal ents” responsible for specific areas of change emotion systems: The “TIES” that form rela- and improvement. For parent–c­hild rela- tionships. Personality and Social Psychology tionships, despite a huge cottage industry Review, 15, 367–393. Butler, E. A., Egloff, B., Wilhelm, F. H., Smith, N. C., Erickson, E. A., & Gross, J. J. (2003). The social consequences of expressive suppres- sion. Emotion, 3, 48–67.

280 SOCIAL ASPECTS Butner, J., Diamond, L. M., & Hicks, A. M. competence. Journal of Child Psychology and (2007). Attachment style and two forms of Psychiatry, 34, 715–728. affect coregulation between romantic part- Doss, B. D., Rhoades, G. K., Stanley, S. M., & ners. Personal Relationships, 14, 431–455. Markman, H. J. (2009). The effect of the tran- sition to parenthood on relationship quality: Campos, J. J., Walle, E. A., Dahl, A., & Main, A. An 8-year prospective study. Journal of Per- (2011). Reconceptualizing emotion regulation. sonality and Social Psychology, 96, 601–619. Emotion Review, 3, 26–35. Dumas, J. E., LaFreniere, P. J., & Serketich, W. J. (1995). “Balance of power”: A transactional Carstensen, L. L., Gottman, J. M., & Levenson, analysis of control in mother–­child dyads R. W. (1995). Emotional behavior in long- involving socially competent, aggressive, and term marriage. Psychology and Aging, 10, anxious children. Journal of Abnormal Psy- 140 –149. chology, 104, 104–113. Erikson, E. H. (1950). Childhood and society. Carstensen, L. L., Graff, J., Levenson, R. W., & New York: Norton. Gottman, J. M. (1996). Affect in intimate rela- Erikson, E. H. (1982). The life cycle completed: tionships: The developmental course of mar- A review. New York: Norton. riage. In C. Magai & S. H. McFadden (Eds.), Erikson, E. H., Erikson, J., & Kivnick, H. Handbook of emotion, adult development, (1986). Vital involvement in old age. New and aging (pp. 227–247). San Diego, CA: Aca- York: Norton. demic Press. Ersner-H­ ershfield, H., Mikels, J. A., Sullivan, S. J., & Carstensen, L. L. (2008). Poignancy: Carstensen, L. L., Isaacowitz, D. M., & Charles, Mixed emotional experience in the face of S. T. (1999). Taking time seriously: A theory meaningful endings. Journal of Personality of socioemotional selectivity. American Psy- and Social Psychology, 94, 158–167. chologist, 54, 165–181. Feeney, J. A. (1999). Adult attachment, emo- tional control, and marital satisfaction. Per- Caspi, A., Henry, B., McGee, R. O., Moffitt, sonal Relationships, 6, 169–185. T. E., & Silva, P. A. (1995). Temperamental Fernald, A. (1991). Prosody in speech to children: origins of child and adolescent behavior prob- Prelinguistic and linguistic functions. Annals lems: From age three to fifteen. Child Devel- of Child Development, 8, 43–80. opment, 66, 55–68. Ferrer, E., & Nesselroade, J. R. (2003). Model- ing affective processes in dyadic relations via Charles, S. T., & Carstensen, L. L. (2007). Emo- dynamic factor analysis. Emotion, 3, 344– tion regulation and aging. In J. J. Gross (Ed.), 360. Handbook of emotion regulation (pp. 307– Fisher, H. E., Aron, A., & Brown, L. L. (2006). 327). New York: Guilford Press. Romantic love: A mammalian brain system for mate choice. Philosophical Transactions of Christensen, A. (1987). Detection of conflict pat- the Royal Society B: Biological Sciences, 361, terns in couples. In K. Hahlweg & M. J. Gold- 2173–2186. stein (Eds.), Understanding major mental dis- Fredrickson, B. L., & Levenson, R. W. (1998). order: The contribution of family interaction Positive emotions speed recovery from the research (pp. 250–265). New York: Family cardiovascular sequelae of negative emotions. Process Press. Cognition and Emotion, 12, 191–220. Gable, S. L., Gonzaga, G. C., & Strachman, A. Coan, J. A., Schaefer, H. S., & Davidson, R. J. (2006). Will you be there for me when things (2006). Lending a hand: Social regulation of go right?: Supportive responses to positive the neural response to threat. Psychological event disclosures. Journal of Personality and Science, 17, 1032–1039. Social Psychology, 91, 904–917. Goodkind, M. S., Gyurak, A., McCarthy, M., Cole, P. M., Teti, L. O., & Zahn-­Waxler, C. Miller, B. L., & Levenson, R. W. (2010). (2003). Mutual emotion regulation and the Emotion regulation deficits in frontotemporal stability of conduct problems between pre- lobar degeneration and Alzheimer’s disease. school and early school age. Development and Psychology and Aging, 25, 30–37. Psychopathology, 15, 1–18. Gottman, J. M. (1993). A theory of marital dis- Côté, S., Gyurak, A., & Levenson, R. W. (2010). The ability to regulate emotion is associated with greater well-being, income, and socioeco- nomic status. Emotion, 10, 923–933. Cowan, C. P., & Cowan, P. A. (1992). When partners become parents. New York: Basic Books. Denham, S. A. (1993). Maternal emotional responsiveness and toddlers’ social-e­motional