Emotion Regulation in Couples Across Adulthood

Abstract

Intimate relationships are hotbeds of emotion. This article presents key findings and current directions in research on couples’ emotion regulation across adulthood as a critical context in which older adults not only maintain functioning but may also outshine younger adults. First, I introduce key concepts, defining qualities (i.e., dynamic, coregulatory, bidirectional, bivalent), and measures (i.e., self-report versus performance-based) of couples’ emotion regulation. Second, I highlight a socioemotional turn in our understanding of adult development with the advent of socioemotional selectivity theory. Third, I offer a life-span developmental perspective on emotion regulation in couples (i.e., across infancy, adolescence and young adulthood, midlife, and late life). Finally, I present the idea that emotion regulation may shift from “me to us” across adulthood and discuss how emotion regulation in couples may become more important, better, and increasingly consequential (e.g., for relationship outcomes, well-being, and health) with age. Ideas for future research are then discussed.

THE GIFT OF AGING?

We recently were handed a precious gift. The human life span expanded in unprecedented ways. A baby born somewhere in the world in 1950 could be expected to live about 46 years. A baby born today has an average life expectancy of 73 years. Around half of the babies born in the 2000s in countries with long-lived populations are predicted to see their one-hundredth birthday (Christensen et al. 2009).

Not only has this gift of an extended life span not been shared equally—there are enormous disparities across geographic regions; racial, ethnic, and socioeconomic backgrounds; and gender and sexual orientations, to name only a few (Thomas Tobin et al. 2023)—but also longer lives have not always been viewed as a gift. Rather, the story of aging has often been told as a story of decline and loss (Carstensen 2011).

Negative views of aging are alive and well today and have, in fact, increased over the past 200 years (Ng et al. 2015). Humans do indeed experience loss and decline across a number of life domains as they age. Physical health and abilities may decline (Rowe & Kahn 1997), the ability to meet basic needs may diminish (World Health Organ. 2020), memory functioning may decrease (Park & Bischof 2013, Salthouse 2004), dementia risk rises (Hou et al. 2019), earlier sources of income and meaning (e.g., career) may be lost, and social networks may shrink (Wrzus et al. 2012).

At the same time, a large body of empirical research shows that individuals maintain or even increase emotional well-being as they get older (e.g., Carstensen et al. 2011, Charles & Carstensen 2010, Charles et al. 2023, Mroczek & Kolarz 1998). Even in contexts that pose objectively greater threats to older adults, such as the COVID-19 pandemic, older adults experience similar or greater levels of positive emotions and similar or lower levels of negative emotions compared with younger adults (e.g., Carstensen et al. 2020, Sun & Sauter 2021, Young et al. 2021).

Pioneering theories of life-span development propose that older adults experience greater emotional well-being because their capacities to regulate emotions are not only spared from decline but also remain stable or increase with age (Carstensen 2006, Carstensen et al. 1999, Charles 2010). This view has received empirical support across a sizable body of work (e.g., Reed et al. 2014). Yet, other studies have yielded nil or negative results, leading some scholars to conclude that empirical evidence currently does “not clearly support the assertation that older adults are better at emotion regulation” (Isaacowitz 2022, p. 1541; see also Allen & Windsor 2019). How can we reconcile these different perspectives?

Research across life-span developmental, affective, and relationship science suggests that, when it comes to emotion regulation in close relationships specifically, older adults may outshine younger adults (Carstensen et al. 1995, 1999; Gottman & Gottman 2017; Levenson et al. 2013; Luong et al. 2011; Verstaen et al. 2020). I focus here on emotion regulation in romantic couples as particularly salient close relationships for many people across cultures (cf. Jankowiak & Fischer 1992). It remains to be seen whether similar developmental trends also emerge for emotion regulation in other close relationships, for example, with friends, children, or grandchildren (Haase 2023). In this article, I review key findings and current directions in research on emotion regulation in couples across adulthood. I highlight important concepts, defining qualities, and measurement approaches; discuss important theoretical frameworks; trace development across the life span; and review evidence suggesting that emotion regulation in couples becomes more important, better, and increasingly consequential with age. I conclude with directions for future research.

CONCEPTS, DEFINING QUALITIES, AND MEASURES

Emotion and Emotion Regulation

The question of what an emotion is has long occupied scholars and laypeople alike (James 1884). Emotion stems from the Latin word “emovere” (from “ex” and “movere,” meaning “out of” and “moving,” respectively), and emotions can be defined as “short-lived psychological-physiological phenomena that represent efficient modes of adaptation to changing environmental demands” (Levenson 1999, p. 481). From this perspective, emotions manifest in emotional behavior (e.g., smiling), voice (e.g., laughter), autonomic physiology (e.g., lower blood pressure), subjective emotional experience (e.g., feeling happy), language (e.g., positive emotion words), and other response systems (Levenson et al. 2017).

Emotions are deeply embedded in the social fabric of our lives. In fact, the word emotion itself emerged with a decidedly sociopolitical connotation in fifteenth-century France (Hochner 2016). Functionalist perspectives view social functions of emotions as critical (Keltner & Haidt 1999, Keltner et al. 2019), and, from this perspective, prototypical emotion antecedents (e.g., loss for sadness, injustice for anger) are often deeply interpersonal (Levenson 1999).

It may come as no surprise that most of our emotion regulation episodes occur in social contexts (estimates suggest up to 98%; Gross et al. 2006) (English et al. 2017). Yet, existing definitions, measures, and studies have often focused on emotion regulation in individuals (Levenson et al. 2013). Influential definitions of emotion regulation have focused on “the processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” (Gross 1998, p. 275, emphasis added). Important approaches for measuring emotion regulation have relied on individuals filling out questionnaires by themselves, watching emotion-eliciting film clips by themselves, and answering prompts on their mobile phones by themselves (Haase 2023). And the majority of existing studies have examined emotion regulation in individuals. In a review of peer-reviewed empirical articles about emotion regulation in adults and adolescents, Campos and colleagues (2011) found that less than 12% of studies examined emotion regulation in a social context.

In the past decade, important progress has been made in furthering our understanding of emotion regulation in social contexts ranging from the macro to the micro, from group-based emotion regulation (Goldenberg et al. 2016), interpersonal emotion regulation (Zaki & Williams 2013), and extrinsic emotion regulation (Nozaki & Mikolajczak 2020) to emotional coregulation (Butler 2011, Butler & Randall 2013), interpersonal linkage (Timmons et al. 2015), and positivity resonance (Fredrickson 2016). However, couples’ perspectives on emotion regulation are still rare. A PubMed search (see Figure 1) reveals a substantial gap between individual- and couple-focused emotion regulation research, which has widened increasingly over time, with close to 2,000 publications referencing emotion regulation and individuals in 2020 compared with a little more than 100 publications referencing emotion regulation and couples in the same year. Perspectives that put couples’ emotion regulation in a developmental context are even rarer (Levenson et al. 2013).

Figure 1.

Number of publications indexed in PubMed referencing “emotion regulation” and “individuals” (blue) versus “emotion regulation” and “couples” (red) between 2000 and 2020.

Similarly, relationship science frameworks have provided vital insights into the initiation, maintenance, and dissolution of interpersonal relationships (e.g., Finkel et al. 2017) but have rarely adopted a life-span developmental perspective. Couples therapy approaches, which have provided critical insights into the importance of couples’ emotion regulation, have had, at least implicitly, a focus on a rather narrow slice of the life span: midlife (Lebow & Snyder 2022).

Defining Qualities of Emotion Regulation in Couples

Just like emotion regulation in individuals, emotion regulation in couples can be explicit and implicit (Gyurak et al. 2011), it can be successful and unsuccessful, and disentangling emotion reactivity and emotion regulation processes can be difficult. Yet, Levenson and colleagues (2013) propose that emotion regulation in couples has a number of defining qualities that render the emotion regulatory landscape that couples face quite different from the one faced by individuals.

First, emotion regulation in couples is dynamic and iterative. Emotions and emotion regulation attempts by one partner may trigger emotional responses in the other partner, resulting in a complex emotion regulatory dance unfolding over time. The resulting emotion and emotion regulation sequences and patterns can become deeply entrenched over the course of a relationship (e.g., “Here we go again,” “I’ve told you a million times”). One commonly studied dynamic where emotion regulation attempts by one partner become a trigger for emotion regulation in the other partner is the demand–withdraw pattern, which is commonly found in mixed-sex (Christensen & Heavey 1990) as well as same-sex (Holley et al. 2010) couples. This pattern describes a dynamic where one partner demands change (e.g., through nagging, criticizing, and pressuring) in an attempt to modify the relationship while the other partner withdraws (e.g., through changing the subject, avoiding, and the silent treatment) in an attempt to preserve the relationship as is.

Second, emotion regulation in couples is coregulatory. Partners each come with their own emotion regulation goals, challenges, and strategies, and they regulate not only their own emotions but also their partner’s. An ever-growing literature has described phenomena of emotional linkage, synchronization, and resonance that can emerge within couples as a result of these coregulatory processes (Butler & Randall 2013, Fredrickson 2016). For example, when couples interact, their physiological processes, facial expressions, and patterns of neural activation can become synchronized such that, for example, both partners’ heart rates accelerate and decelerate in similar ways or one partner’s heart rate accelerates while the other’s decelerates (Chen et al. 2021).

Third, emotion regulation in couples is bidirectional. While emotion regulation is often aimed at upregulating or maintaining positive emotions and downregulating negative emotions, there are important contexts where people seek to downregulate positive emotions and maintain or upregulate negative emotions (Tamir 2016). One such context is the couple context. Emotions, especially negative emotions, are often shared (and regulated) within couples, and sometimes one’s partner may be the only person with whom one’s emotions are shared (Rime 2009). This can render some emotion regulation strategies that are widely adaptive for individuals quite problematic when used in a couples context. For example, positive reappraisal (i.e., positively reframing thoughts in order to upregulate positive and downregulate negative emotions) is a regulation strategy that comes with a host of intrapersonal benefits (but see Ford & Troy 2019, Hittner et al. 2019). However, this strategy may have interpersonal costs when individuals want to share negative emotions and long for their partner to embrace them without judgment, rather than being told to look on the bright side.

Finally, emotion regulation in couples is bivalent, as partners regulate both negative and positive emotions. The regulation of negative emotions has understandably received a lot of attention given the many consequences of negative emotions for well-being and health (e.g., Haase et al. 2016). However, the regulation of positive emotion may be particularly important in the context of close relationships. Some of our most important positive emotions—love (Fredrickson 2016, Reis & Aron 2008), lust (Panksepp 1998), desire (Diamond 2008)—are deeply interpersonal, and their down- and upregulation have posed conundrums for laypeople and scholars throughout the ages. In this vein, it is critical to note that abominable practices, rooted in homophobia and transphobia, have sought to manipulate exactly those positive emotions through, for example, conversion therapy.

Measures

To study emotion regulation, researchers have often relied on questionnaires, which require individuals to consciously reflect and report on their emotion regulation. One of the most widely used questionnaires, the Emotion Regulation Questionnaire (e.g., Gross & John 2003), measures how individuals “control (that is, regulate and manage)” their emotions through reappraisal, which refers to the reframing of one’s thoughts (e.g., “When I want to feel less negative emotion [such as sadness or anger], I change what I’m thinking about.”), and expressive suppression, which refers to the inhibition of emotional expressive behaviors (e.g., “When I am feeling negative emotions, I make sure not to express them.”). Other questionnaires capture interpersonal emotion regulation, for example, as people recruit social resources to manage their emotions (e.g., “I find that even just being around other people can help me to feel better”; Williams et al. 2018). Few questionnaires measure emotion regulation in couples specifically, but one that does is the couple version of the Interpersonal Emotion Regulation in Close Relationships questionnaire (Horn 2022), which measures codistraction, coreappraisal, positive humor, and physical affection as well as cobrooding, cosuppression, and negative humor within a close relationship.

Self-report measures are easy to administer, allow for large-scale assessments, provide information on individuals’ beliefs about their emotion regulation goals and strategies, and have provided important insights into emotion regulation across adulthood (e.g., Gross & John 2003). Nonetheless, these measures are not without limitations. They cannot capture implicit emotion regulation (cf. Gyurak et al. 2011). Self-presentational and attribution response biases may limit their validity (Robinson & Clore 2002). They cannot capture the dynamic, coregulatory emotional dance unfolding in a couples’ relationship (Levenson et al. 2013). And they are unable to measure emotion regulation across the multiple response systems in which emotions manifest. This is particularly important since people may not necessarily be aware of the effects that emotion regulation strategies can have on themselves and their partners. For example, expressive suppression can reduce emotional behavior, while at the same time increasing physiological arousal in not just the suppressor but also an interaction partner (Butler et al. 2003).

Thus, another line of research has used performance-based measures of emotion and emotion regulation, studying couples as they talk about disagreements in the relationship (e.g., Carstensen et al. 1995) or things they enjoy doing together (e.g., Brown et al. 2022) as well as stressful events (e.g., Meier et al. 2021) or positive events (e.g., Gable et al. 2006) in their lives (Gottman & Gottman 2017, Karney & Bradbury 2020, Kiecolt-Glaser et al. 2005, Levenson et al. 2013). Many of these studies have used laboratory-based approaches to examine emotion regulation across multiple response systems in both partners, including emotional behavior (e.g., using interpersonal behavior coding systems), autonomic physiology (e.g., using cardiac, vascular, electrodermal, and respiratory assessments), subjective emotional experiences (e.g., using emotion checklists or rating dials), and language (e.g., using word frequency analysis). Some studies of emotion regulation in couples have used seemingly simple paradigms and have shown, for example, how holding hands with one’s spouse (versus holding hands with a stranger or being in a solo condition) can have profound emotion regulation effects, reducing neutral activation in threat-relevant areas and subjective distress (Coan et al. 2006). Beyond the laboratory, ambulatory studies have provided important insights into emotion and emotion regulation in couples in everyday life (e.g., Meier et al. 2022), such as by assessing naturalistic conversations and sampling across multiple response systems, including those not easily observable in the laboratory (e.g., activity; Chen et al. 2022).

THEORETICAL FRAMEWORKS

Humans are ultrasocial beings, and high-quality close relationships are important for well-being across cultures (e.g., Demir & Sümer 2018, Proulx et al. 2007). The idea that “a person is a person through a person” and that we are all interconnected is at the heart of Ubuntu philosophy and features in religious, philosophical, sociological, and psychological frameworks across ages and cultures (e.g., Baumeister & Leary 1995, Frankl 1963, Gottman & Gottman 2017, Rosa 2019). Some of these frameworks emphasize the importance of being in relationship when faced with existential questions in particular (Frankl 1963). These questions, arguably, become more salient as we approach the end of life.

Earlier developmental scholars did not view social relationships as particularly important in late life (Haase & Shiota 2019). Rather, they viewed late life as a time to disengage and prepare for death (Cumming & Henry 1961), as a time to adjust to physical, mental, social, and economic limitations (Havighurst 1976), as a time to turn inward and flatten emotions (Jung 1933), and as a time of increased focus on the self (Erikson 1950).

This changed with the advent of socioemotional selectivity theory (e.g., Carstensen 2006, Carstensen et al. 1999). Socioemotional selectivity theory put social relationships at the center of our understanding of late-life development, highlighting in particular the emotion regulation functions of close relationships. As people age, the theory posits, they become increasingly aware that their time on this earth is limited and prioritize emotional well-being and close relationships in the here and now. They invest less time and effort in pursuing knowledge-related goals that promise to pay off in the long term (e.g., dating, career explorations) but can produce negative emotions in the short term. Instead, as people grow older, they are thought to increasingly focus on socioemotional goals that maximize emotional well-being in the here and now, including affectionate and enjoyable interactions within close relationships.

The model of strength and vulnerability integration (SAVI) (Charles 2010) similarly centers emotion regulation as a key to understanding well-being trajectories across adulthood. The model proposes that adults become better at emotion regulation with age, specifically when it comes to using strategies to avoid or limit exposure to negative stimuli. One context in which older adults show notable strengths is close relationships. At the same time, when they cannot avoid or reduce exposure to negative stimuli, older adults are thought to show vulnerabilities and experience difficulties with emotion regulation. Socioemotional selectivity theory, the SAVI model, and other influential frameworks (e.g., Urry & Gross 2010) have sparked an enormous body of empirical research aiming to understand emotion regulation across the adult life span alongside other fields that have seen a rise in affectivism (Dukes et al. 2021).

A LIFE-SPAN DEVELOPMENTAL PERSPECTIVE ON EMOTION REGULATION IN COUPLES

There is substantial evidence that individuals become better at emotion regulation with age (Charles & Carstensen 2010, Urry & Gross 2010). Recent studies supporting this view come, for example, from the COVID-19 pandemic. Studies not only show that older adults, on average, fared considerably better in the pandemic than younger adults in many countries across the globe, despite having objectively greater health risk (e.g., Carstensen et al. 2020, Sun & Sauter 2021, Young et al. 2021). They also demonstrate that age-related advantages in emotional well-being can, at least in part, be explained by greater use of adaptive and lower use of maladaptive emotion regulation strategies among older adults (Dworakowski et al. 2022, Young et al. 2021). At the same time, laboratory-based research has revealed a more nuanced picture, showing, for example, age-related advantages in the ability to implement positive reappraisal but age-related decline in the ability to implement detached reappraisal (i.e., adopting an unemotional and neutral perspective) (Shiota & Levenson 2009).

Inspecting the available empirical evidence has led some scholars to conclude that existing studies “do not clearly support the assertion that older adults are better at emotion regulation” (Isaacowitz 2022, p. 1541). Similarly, Allen & Windsor (2019, p. 1) have stated that “adult development is not characterised by straightforward shifts in preferences for use of different [emotion regulation] strategies.”

Perhaps, rather than increasing across the board, older adults are better at emotion regulation in specific contexts. One context in which older adults really shine may be the couples context as emotion regulation changes from “me to us” across the life span (Haase 2023). Before highlighting the key ideas behind this proposition and supporting empirical evidence, it is important to consider the specific challenges, opportunities, constraints, and affordances for emotion regulation in couples across the life span, beginning in infancy.

Infancy

For all of us, emotion regulation begins as a “we” at the very beginning of our lives. Fetuses in the womb, for example, make more cry-face expressions when the mother eats bitter food and more laughter-face expressions when the mother eats nonbitter food (Ustun et al. 2022). These findings not only highlight the deep emotional interconnectedness between fetus and mother, but together with many other pieces of empirical evidence (e.g., Cowen et al. 2021, Tracy & Matsumoto 2008, Valente et al. 2018; but see Barrett 2014), they also support the idea that there are indeed some emotions that have evolved to solve specific evolutionary challenges—and that people then often seek to modulate through emotion regulation (Levenson 1999).

When they are born, infants remain deeply emotionally dependent on their caregivers. At the same time, the reverse may also apply. Infants regulate their caregivers’ emotions in profound ways (Tronick 1989). Their smiles may become highly cherished rewards for their caregivers. Their crying may prompt elaborate soothing, rocking, singing, feeding, and changing rituals. These coregulatory emotional dances (Feldman 2007, Waters et al. 2014) are at the heart of infant–caregiver attachment across diverse cultures (Causadias et al. 2022, Mooya et al. 2016). Attachment theory posits (Bowlby 1988) that these early experiences then serve as a blueprint for how individuals navigate close relationships decades later in life (Shaver & Mikulincer 2007, Sroufe et al. 2010).

Adolescence and Young Adulthood

Western scholars have sometimes described adolescence and young adulthood as a time of the “developmental me” (Roberts et al. 2010). Socioemotional selectivity theory proposes that the pursuit of knowledge-related goals takes center stage (Carstensen et al. 1999), and other frameworks similarly emphasize the importance of exploration (and specifically identity exploration) in adolescence and young adulthood (Arnett 2000, Erikson 1950). These life stages are times of profound change across emotional, neurobiological, motivational, cognitive, and social systems (Silvers 2022, Silvers & Peris 2023). While there is important cultural variation (Qu & Telzer 2018), adolescents across a variety of species show behavioral changes that have evolved to support independence (Spear 2000). Social relationships are deeply important in adolescence and young adulthood (e.g., Rodosky et al. 2023), but young people may seek different kinds of relationships for somewhat different reasons than older adults. They may seek to expand their social networks rather than focus on close others (English & Carstensen 2014, Shaver & Mikulincer 2007). Their social relationships may serve exploration goals rather than well-being goals (Carstensen et al. 1999). Their skills to emotionally navigate these relationships may just be beginning to mature (Silvers 2022). And they may engage in profound exploration when it comes to romantic relationships in particular.

Romantic relationships can begin at any time, but young adulthood is often thought of as the prime time of relationship beginnings. Early romantic love can have intoxicating experiential, behavioral, physiological, and neurobiological qualities (Fisher et al. 2006). Metaphors (Lakoff & Johnson 1980) in English and other languages (e.g., falling in love, tomber amoureux, sich Hals über Kopf verlieben, muerto de amor) point to one of many emotion regulation conundrums as individuals may perceive having preciously little control over their emotions in this stage. Another conundrum reminds us of the bivalent nature of couples’ emotion regulation as individuals may experience both high highs—feelings of excitement, elation, and euphoria, extreme focus on the desired partner, being full of energy and having trouble sleeping (Fisher et al. 2006)—and low lows, perhaps because of the enormous uncertainty of this phase (see also Silvers & Peris 2023).

Midlife

Midlife is an important time for the development of emotion regulation in couples. Much as children shift from cognitive exploration to exploitation as they grow into adulthood (Gopnik 2020), similar shifts may happen when it comes to couples’ emotion regulation as people move into middle age. While social exploration is a hallmark of adolescence and young adulthood, midlife brings with it a host of social commitments, many of which are quite hard to dissolve and some almost impossible (e.g., relationships with children or aging parents). Thus, people may increasingly refine their emotion regulation skills to modify the social relationships they are in.

Midlife has received surprisingly little scientific attention compared with other life stages (Infurna et al. 2020, Lachman 2015). Yet, many studies of emotion regulation in couples center middle-aged couples (often without highlighting this developmental stage) and paint a general picture of negative emotions as maladaptive and positive emotions as beneficial for relationship quality and stability and a host of other outcomes (e.g., Gottman & Gottman 2017, Karney & Bradbury 1995, Levenson et al. 2013). Yet, a closer look reveals a far more nuanced picture in terms of how specific negative and positive emotions and emotion regulation strategies predict relationship outcomes concurrently and longitudinally (e.g., Bloch et al. 2014, Dubé et al. 2023). For example, greater difficulty downregulating negative emotions during a sexual conflict conversation in married couples is associated with lower sexual well-being concurrently, while longitudinal associations are less clear (Dubé et al. 2023).

The picture becomes even more nuanced when considering the many consequences of couples’ emotion regulation not only for relationship functioning but also for well-being, health, and even longevity (Kiecolt-Glaser & Newton 2001, Smith et al. 2014, Wells et al. 2022). Some negative emotions (e.g., anger), for example, may be quite adaptive in midlife relationships when they propel partners toward active engagement and problem solving (Haase et al. 2012, Karney & Bradbury 2020, Kunzmann et al. 2014) while they, at the same time, may put strain on the cardiovascular system (Meier et al. 2023b) and predict the development of cardiovascular symptoms (Haase et al. 2016). The picture takes on further complexity when considering the many contexts which couples’ relationships are embedded in—across different racial and ethnic, socioeconomic, and cultural backgrounds as well as different sexual and gender orientations (e.g., Karney & Bradbury 2020, Ong et al. 2022, Rogers et al. 2021). Some emotion regulation processes (e.g., reappraisal), for example, may be particularly important for mental health among couples from lower socioeconomic backgrounds, while they may be less central for couples from higher socioeconomic backgrounds (Hittner et al. 2019; see also Troy et al. 2017).

Late Life

Emotion regulation at the level of the “we” becomes central in late life as time horizons shrink and individuals become increasingly selective in their relationships (Carstensen et al. 1999). While couples’ relationships can begin at any point in life, those relationships that have survived the earlier culling of social networks (English & Carstensen 2014) can be very hard to leave and replace, and they have a special place in people’s life stories (McAdams et al. 2022).

Marriages are especially important in late life (93% of US Americans will have been married once in their life by the time they reach age 65, according to the US Census Bureau), and older adults are even more likely to spend time with their spouses than with their friends (Charles & Carstensen 2007). This increased focus on marriage, along with the trimming of peripheral relationships (English & Carstensen 2014), disengagement from life domains (e.g., career) that were important earlier in life (Heckhausen et al. 2010), decreased opportunities to leave dysfunctional relationships, and low divorce rates in late life (Verstaen et al. 2020) make marriages and other committed romantic relationships high-stakes contexts for older adults. When things go well in late-life marriages, older adults can derive pronounced benefits, but when things go awry, marriages can confer notable risks (Charles 2010, Charles & Carstensen 2007, Henry et al. 2007).

Thus, while emotional journeys may be stormy at times for younger couples, couples in late life may prefer smooth sailing and steer toward calmer waters (Kunzmann et al. 2014). And older adults may not only prioritize their own emotional well-being, but may also be highly motivated to invest in the emotional well-being of their partners. This may make it more likely that those partners will return the favor (reciprocal altruism; Trivers 1971) and provide the social support that older adults need during times of loss, decline, and obstacles across multiple life domains (Buchinger et al. 2022).

COUPLES’ EMOTION REGULATION IN LATE LIFE: FROM ME TO US?

Socioemotional selectivity theory and research across life-span developmental, affective, relationship, and neurobiological science provide the backdrop against which we can formulate questions for future research (see Figure 2) about emotion regulation in couples across adulthood (see also Haase 2023).

Figure 2.

Questions for future research about emotion regulation in couples across adulthood.

Does Emotion Regulation in Couples Become Increasingly Important Across Adulthood?

Numerous studies show that, with age, individuals overall become increasingly agreeable (Roberts et al. 2006), less narcissistic (Weidmann et al. 2023), and more trusting (Poulin & Haase 2015). They shift from self-focused to ego-transcendent and prosocial goals (Brandtstädter et al. 2010, Buchinger et al. 2022) and show greater prosocial behavior (Sze et al. 2012, Zak et al. 2022). Interestingly, while the exploration of the self and one’s identity are thought to be prime developmental tasks in adolescence and young adulthood, the pursuit of self-focused goals is associated with higher depression risk even in these life stages, and this risk is mitigated when young people also hold goals that center intimate relationships and emotional well-being (Salmela-Aro et al. 2012). Broader motivational shifts from me to us across adulthood (Steltenpohl et al. 2019) are accompanied by a focus on close relationships specifically (English & Carstensen 2014). And even within couples’ relationships, older adults become more we-oriented: Evidence from laboratory-based as well as experience-sampling studies shows that older couples use more “we” words (e.g., we, us, our) when talking with each other than do younger couples (Meier et al. 2022, Seider et al. 2009).

Research Directions

We know from empathy research that motivation plays a critical role in whether individuals dial their empathy up or down (Zaki 2014). Similar processes may be at play when it comes to emotion regulation in couples. Socioemotional selectivity theory proposes—and a mounting body of evidence supports this view—that shrinking time horizons make individuals prioritize close relationships and emotional well-being in the here and now (Charles & Carstensen 2010). Yet, there is a dire need to develop measures that can capture emotion regulation goals specifically in couples and test age differences across adulthood. For example, with age, individuals may increasingly hold prohedonic emotion regulation goals not just for themselves (e.g., “I want to be happy”) (Riediger et al. 2009) but also for their partners (e.g., “I want you to be happy”). Or they may increasingly conceptualize their emotion regulation goals at the level of the couple (e.g., “I want us to be happy”) rather than the individual.

It is also possible that, with age, partners’ emotion regulation goals and strategies become increasingly aligned as a result of a shift from me to us (Haase 2023). Consider, for example, a couple returning home from a long day at work. One partner wants to vent about an injustice and hopes for their partner to respond with understanding and support for their anger and outrage. The other partner, in contrast, wants to relax and wishes their partner could just calm down and feel better and tells them as much. Add children, friends, neighbors, or pets to the equation and a couple’s emotion regulatory dance can easily turn into a ball. Research could examine whether couples experience fewer of these emotion regulation goal and strategy conflicts with age.

Another consequence of an age-related increase in the importance of emotion regulation in couples may be that emotion regulation success is increasingly defined at the level of the couple and not just the individual. This idea could be probed by exploring what people in different age groups consider to be good emotion regulation, for example. In a broader vein, we may need to adopt more nuanced perspectives on emotion regulation strategies (e.g., positive reappraisal) that are intrapersonal success stories but that might have emotional costs for relationship partners (Ford & Troy 2019)—perhaps especially so in late life (Rompilla et al. 2022). Conversely, we could temper our view of emotion regulation strategies (e.g., expressive suppression) that have intrapersonal costs by considering that these can be very adaptive in some relationship contexts—for example, when caring for a loved one with a neurodegenerative disease in late life (Wells et al. 2021).

Does Emotion Regulation in Couples Become Increasingly Better Across Adulthood?

Earlier cross-sectional studies provided important insights into age-related differences in couples’ emotion regulation, showing, for example, that conflict resolution was less emotionally negative and more affectionate for older couples compared with middle-aged couples (Carstensen et al. 1995). In a similar vein, longitudinal studies have now shown that couples become better at emotion regulation with age. In our own work, we found overall decreases in spouses’ negative interpersonal behaviors and overall increases in spouses’ positive interpersonal behaviors during a 15-minute marital conflict conversation in a sample of long-term middle-aged and older married couples over more than a decade (Verstaen et al. 2020).

In this study, interpersonal behaviors were coded on a second-by-second basis using the Specific Affect Coding System (Coan & Gottman 2007), which is based on a gestalt of facial expressions, gestures, body postures, tones of voice, and speech contents, which allowed for tracking changes in specific multiple positive and negative behaviors over time. Belligerence, defensiveness, fear or tension, whining, and stonewalling (for men) and contempt (for women) decreased with age, while other negative emotional behaviors, notably sadness (cf. Haase et al. 2012, Kunzmann et al. 2014), did not. Moreover, converging with important research on age differences in emotional acceptance in individuals (Wolfe & Isaacowitz 2022), this study (Verstaen et al. 2020) showed increases in spouses’ acceptance (labeled validation) behavior across time alongside increases in spouses’ humor and enthusiasm. That is, with age, spouses not only used more good-natured humor, jokes, and silliness and expressed more joy, but also increasingly showed behaviors that communicated sincere understanding and acceptance of one’s partner and their emotions and thoughts.

Another longitudinal study, drawing from the same data set, showed increases in avoidance behavior during marital conflict with age (Holley et al. 2013), with spouses increasingly avoiding discussions of the problem by hesitating, changing topics, diverting attention, or delaying the discussion. These findings converge with a sizable body of research on emotion regulation in couples specifically and social relationships more broadly (Luong et al. 2011), which shows that older adults increasingly engage in strategies that optimize positive social experiences and minimize negative ones.

Research Directions

Research has begun to document age-related changes in emotion regulation strategies that individuals use in couples contexts using longitudinal designs, and some of them have involved multiple cohorts (Verstaen et al. 2020). Such studies will be critical in advancing our understanding of dyadic emotional aging, especially when they allow for disentangling age and cohort effects. Careful design choices (such as studying long-term relationships with low divorce risks, choosing follow-up windows that allow for detecting meaningful changes; e.g., Verstaen et al. 2020) and statistical decisions (e.g., probing and accounting for selective drop-out; Hoppmann et al. 2011) can go a long way toward ruling out alternative explanations for age-related changes. Moreover, faced with the conundrum of how to disentangle the effects of age and relationship length, future research could not only examine relationship length as a moderator but also probe emotion regulation in interesting couple contexts (e.g., younger versus older dating or newlywed couples).

Moreover, much like the majority of relationship research, past studies have relied on “mostly White, heterosexual couples” (McGorray et al. 2023). It is critical to move closer to an inclusive science that examines age-related changes in couples’ emotion regulation abilities and practices across different relationship types, gender and sexual orientations, racialized and ethnic groups, socioeconomic backgrounds, and cultures.

Finally, we need studies to probe whether a shift from me to us occurs at the level of couples’ emotion coregulation (e.g., Butler 2011, 2015) with age. Existing studies, for example, have elucidated how partners’ emotions become synchronized or desynchronized in the laboratory and daily life and how important these coregulation processes are for relationship quality and spouses’ health (e.g., Chen et al. 2022, Wells et al. 2022). Yet, we do not know how couples’ emotion coregulation changes with age or, more simply, whether and how younger and older couples differ with respect to the degree and kind of linkage across different response systems.

A particularly fruitful avenue may be to examine age-related changes in shared positive emotions—positivity resonance—which has emerged as a critical building block of relationship satisfaction, well-being, health, and even longevity (Fredrickson 2016, Otero et al. 2019, Wells et al. 2022). Positivity resonance describes moments of interpersonal connection that emerge when two or more individuals simultaneously experience positive emotions that are elevated by the presence of key behavioral and physiological features. Positivity resonance can be measured using questionnaire-based measures of perceived positivity resonance (e.g., “did you feel in sync with the other[s]?”) as well as using performance-based measures at the experiential, behavioral, and physiological level as we have done recently (Wells et al. 2022). There exist several dyad-level behavioral coding systems, for example, that capture positivity resonance, including behavioral indicators of positivity resonance (e.g., do partners’ actions, words, or voice intonation convey mutual warmth, mutual concern, mutual affection, and/or a shared tempo?). Whether positivity resonance within couples or other close relationships increases with age is an exciting question for future research.

Does Emotion Regulation in Couples Become Increasingly Consequential Across Adulthood?

How couples regulate emotions has important consequences for relationship quality and stability (Bloch et al. 2014, Gottman & Gottman 2017, Karney & Bradbury 1995); for intimate partner violence (Maloney et al. 2023); for health behaviors such as sleep, substance use, or eating (e.g., Butler et al. 2010); and for mental health, physical health, and longevity (e.g., Haase et al. 2016, Kiecolt-Glaser & Wilson 2017, Smith et al. 2014).

There is some evidence that these links may become even stronger in late life. For example, whether individuals perceive negative or positive emotional behaviors in their spouses appears to be an even stronger predictor of marital quality in late life compared with midlife (Henry et al. 2007). Similarly, meta-analytic evidence shows that the link between attachment avoidance and lower relationship satisfaction becomes even more pronounced with age (Candel & Turliuc 2019). While it remains to be seen to what extent alterations in dyadic emotion regulation drive these effects, it is interesting to note that attachment avoidance is characterized by an avoidance of emotional “we” states in relationships. People with higher avoidant attachment may seek to “avoid getting too close” and “prefer not to show a partner how [they] feel deep down”; they may not be “comfortable sharing their private thoughts and feelings with their partner” and may have a hard time turning to their partner “for many things, including comfort and reassurance” (all example items from a commonly used scale to measure attachment avoidance in adults; Brennan et al. 1998).

Longitudinal studies have been particularly important in demonstrating some of the far-reaching developmental consequences of couples’ emotion regulation (e.g., Bloch et al. 2014, Chen et al. 2021). A recent study (Wells et al. 2022), for example, focused specifically on positivity resonance as an important aspect of emotion (co)regulation in close relationships. Drawing from Fredrickson’s (2016) positivity resonance theory and the longitudinal study of long-term marriages mentioned above (e.g., Verstaen et al. 2020), findings showed that positivity resonance during marital conflict predicted greater health and longevity over 30 years. Importantly, the effect on longevity held when controlling for a host of covariates, including income, education, health behaviors, marital satisfaction, and individuals’ positive emotions during the conversation, thus emphasizing that emotion regulation for “us” is more than the sum of emotion regulation for each individual “me.”

The couples therapy literature provides further evidence supporting the importance of couples’ emotion regulation not just for relationship quality and stability but also for individual well-being and health. Couples therapy approaches that center emotion and emotion regulation include emotionally focused couples therapy (Johnson 2019) and Gottman Method Couples Therapy (Gottman & Gottman 2015); emotions also play a critical role in cognitive-behavioral couples therapy and couple-based interventions for psychopathology (Baucom et al. 2019), among other approaches.

These couples therapy approaches have long been shown to be effective not only in ameliorating relationship distress (Garanzini et al. 2017, Spengler et al. 2022) but also in improving individuals’ mental health (Baucom et al. 2014, Knobloch-Fedders et al. 2015, Shapiro & Gottman 2005), and there is some evidence for physiological benefits as well (e.g., Ditzen et al. 2011). Although more research is needed, increases in partners’ emotion regulation skills have been suspected to be one of the active ingredients that mediate the beneficial effects of couples therapy (Baucom et al. 2019). For example, in a pilot study, Kirby & Baucom (2007) showed that teaching emotion regulation in a yearlong dialectical behavior therapy skills training group with couples in which at least one partner had experienced chronic emotion regulation difficulties led to increases in adaptive emotion regulation and alleviated depressive symptoms.

Research Directions

Couples’ emotion regulation matters for developmental outcomes, from relationship satisfaction and stability to well-being and mental and physical health. Given the lack of research in this area (see Figure 1), more studies are urgently needed, particularly with couples that have often been marginalized in research, and perhaps especially so looking at critical late-life outcomes, such as physiological reactivity (Gordon & Mendes 2021) and physical health (e.g., Kiecolt-Glaser & Wilson 2017, Pietromonaco & Collins 2017). Even in the absence of statistical moderation, emotion regulation in couples is particularly consequential in late life precisely because couples’ relationships are so important in this life stage and because there is, simply put, less time to correct course and start a new relationship.

Future research has a lot to gain by entering into deeper conversation with couples therapy researchers and practitioners. Models of couples therapy have implicitly centered on mid-life couples (Lebow & Snyder 2022), and interventions in general have often been regarded as the province of earlier life stages. Yet, older couples often face existential challenges—medical decisions, caregiving, illness and death of loved ones, and end-of-life decisions, to name a few—that put their emotion regulation capacities to the test. Education and therapy with older couples that center emotion regulation could provide unique insights into the plasticity of emotion regulation later in life, and they promise to yield important benefits even beyond the couple (e.g., for adult children).

We also need to better understand the underlying mechanisms that may render couples’ emotion regulation so consequential in late life. Increases in positive emotions and decreases in threat reactivity are two possible candidate mechanisms. Negative emotions have received lots of scientific attention (cf. Algoe 2019, Shiota et al. 2017), but positive emotions are important across the life span and perhaps especially so in late life (see e.g., Hittner et al. 2020, Pressman et al. 2019, Wells et al. 2022) as they serve as important physiological repair mechanisms (Fredrickson & Levenson 1998, Fredrickson et al. 2000). Showing joy and enthusiasm (i.e., capitalization) and validation, caring, and understanding (i.e., responsiveness), for example, have been shown to protect older couples from shared physiological stress (Shrout et al. 2023). Determining how couples coregulate positive emotions (e.g., through responsiveness, capitalization, or humor), which positive emotions they long for (e.g., calm or excitement), what contexts these positive emotions can be created in (cf. Moskowitz et al. 2019), and how these positive emotion processes benefit aging-relevant outcomes are all important questions for future research.

Moreover, neuroscientific research grounded in social baseline theory (Coan & Sbarra 2015) raises the possibility that couples’ emotion regulation reduces threat reactivity, which in turn could promote adaptive aging. Some frameworks posit that one ultimate source of threat in late life is an increased awareness of the finitude of one’s life (Becker 1973), and close relationships may be a prime avenue through which individuals may keep this threat at bay. Not only does the physical proximity of close relationship partners (e.g., through hand holding) reduce threat (Coan et al. 2006), but also close relationship partners may literally get under one’s skin and become neurally embedded in the self, thus buffering threat (Beckes et al. 2013). The embedding of the individual “me” into a relational “us” could promise to be a way to live on in the heart, mind, and brain of a loved one. Some initial evidence supporting this idea comes from a recent 20-year longitudinal study, which showed that greater interdependence predicted greater longevity over and above covariates, adding, on average, about two years of life (Ng et al. 2020). Importantly, this association was mediated fully by greater serenity toward death.

Finally, even those emotion regulation processes that serve a couple very well can have maladaptive consequences and come with sizable risks down the road. These could include increased negative emotions toward outgroup members who are not part of the we (De Dreu et al. 2011) and a cascade of vulnerabilities after the death of one’s partner (Meier et al. 2023a), which could be particularly devastating for surviving partners who have merged more closely emotionally with their partner. It will be important to test and keep the consequences of couples’ emotion regulation in all their complexity in mind.

CONCLUSION

If things go well, we will all grow old. However, laypeople, policy makers, and developmental scholars have often had a rather bleak outlook on aging. Research on emotion regulation across adulthood shows that there is good reason for hope, especially when it comes to emotion regulation in close relationships and particularly within couples. With age, emotion regulation in couples may become more important, people may become better at it, and it may become increasingly consequential for developmental outcomes ranging from relationship quality to health. More research on couples’ emotion regulation across the life span is needed to test these ideas, in line with the important original theoretical propositions (Carstensen et al. 1999, Charles 2010) and reflecting the social embeddedness of emotion regulation (Levenson et al. 2013).

It will be particularly important to pay close attention to how couples’ emotion regulation dynamics play out for couples across racial and ethnic identities, socioeconomic backgrounds, and cultures as well as gender and sexual orientations. Important work remains to be done to increase our understanding of gender differences that have repeatedly emerged in couples research (e.g., Kiecolt-Glaser & Newton 2001). Quantitative (e.g., Holley et al. 2010, Ong et al. 2022) as well as qualitative (e.g., Hammack et al. 2019) approaches will both be important in augmenting and transforming existing research in this area. Moreover, cultural beliefs, norms, and values deserve greater attention. Studies, not just on emotional development, have often focused on samples from global minority cultures (Western, educated, industrialized, rich, democratic) that value independence and autonomy (Henrich et al. 2010). We need more research including samples from global majority cultures that prioritize interdependence and connectedness. Longitudinal studies will be particularly important as increases in individualism have now been documented in recent cohorts across more than 70 countries (Santos et al. 2017).

Finally, a focus on emotion regulation within couples specifically and social relationships broadly may help us appreciate that emotions are important not only for individual development but also for the development of couples, communities, and cultures. As bell hooks (1994, p. 247) reminds us, “it is in choosing love, and beginning with love as the ethical foundation for politics, that we are best positioned to transform society in ways that enhance the collective good.”