Abstract
Compared with securely attached people, insecurely attached people have romantic relationships that are less happy and more unstable, but the quality of their relationships should depend on how their partners regulate them. Some partners find ways to buffer (emotionally and behaviorally regulate) insecurely attached individuals, which helps such individuals feel better and behave more constructively and improves the relationship. Understanding when and how this important interpersonal process works requires a dyad-centered approach. The present research describes core tenets of attachment theory and the two forms of attachment insecurity (anxiety and avoidance) and presents our dyadic regulation model of insecurity buffering, which explains how and why certain types of buffering behaviors soothe the worries and improve the relationship perceptions and behaviors of anxious or avoidant people. Studies that illustrate ways in which partners can successfully buffer the insecure reactions of anxious and avoidant individuals are reviewed, and other traits and social contexts to which our model can be applied are also discussed.
Matthew and Helen have a long-standing romantic relationship. Because of bad experiences with prior partners, Matthew worries that Helen does not completely love him. These chronic worries make Matthew occasionally “act out,” especially when he and Helen have major disagreements and Matthew believes the relationship could be threatened. Helen, however, has learned to give Matthew the reassurance he needs to control his emotions, feel more secure about the relationship, and behave more constructively when conflicts arise. Over time, Helen’s actions have helped Matthew feel better about himself and the relationship, and the two are now happier than ever.
This fundamental interpersonal process, called partner buffering, occurs every day in relationships. One might assume that partner buffering has been widely studied and is well understood, but neither is the case. Instead, most researchers have adopted an individual-centered viewpoint to examine how Matthew’s insecurity affects his relationship functioning. A complete understanding of partner buffering, however, requires a dyad-centered approach to determine what Matthew’s partner (Helen) does to allay Matthew’s insecurity and help him to feel and behave better. In the research presented here, we introduce a dyadic regulation model of insecurity buffering and then describe behavioral observation studies of couples engaged in conflict discussions that illustrate ways in which partners can buffer two types of insecurity in relationships.
Attachment Insecurity and Relationship Functioning
According to attachment theory (Bowlby, 1969, 1973), humans evolved to form strong emotional bonds with their caregivers (attachment figures) because doing so increased the probability of survival. These attachment bonds operate in children and adults, especially when individuals feel threatened, distressed, or challenged (Bowlby, 1969; Simpson & Rholes, 2012). Depending on how they have been treated by prior attachment figures (e.g., parents, close friends, romantic partners), individuals develop different ways of viewing and relating to their current attachment figures, known as attachment orientations. Securely attached people have received good care and support from prior attachment figures, which leads them to have positive views of themselves and others and motivates them to turn to their attachment figures for comfort and support to reduce negative affect and remove the source of distress. This collaborative, “problem-focused” style of coping helps securely attached people build further closeness and intimacy with their partners (Mikulincer & Shaver, 2007).
There are two primary types of insecurity. Anxiously attached people, who have received mixed or inconsistent care/support earlier in life, crave acceptance and closeness yet worry that their partners might hurt or leave them (Mikulincer & Shaver, 2003). Consequently, anxious individuals are vigilant to signs of both their partner’s love and their partner’s possible rejection, which generates strong distress and dysfunctional behavior in relationship-threatening situations, such as major relationship conflicts (Simpson, Rholes, & Phillips, 1996) or when partners do not provide sufficient support (Collins & Feeney, 2000). This “hypervigilant” form of coping allows anxious individuals to monitor and gain the attention of their partners, which makes them feel less insecure. Unfortunately, these responses typically produce anger and dissatisfaction in their partners (Downey, Freitas, Michaelis, & Khouri, 1998).
Avoidantly attached people, who have been rejected or dismissed by earlier attachment figures, believe they cannot trust and depend on others. Accordingly, they learn to suppress their needs for closeness and intimacy and become rigidly independent and self-reliant. When stressed, avoidant individuals withdraw from their partners emotionally (Pietromonaco & Feldman Barrett, 1997) and become less inclined to seek or give support (Simpson, Rholes, & Nelligan, 1992). This “distancing” strategy allows avoidant people to maintain sufficient autonomy and independence so they can regulate their emotions and handle the source of distress on their own (Mikulincer & Shaver, 2003).
Both types of insecurity destabilize relationships by lowering satisfaction, aggravating relationship problems, and curtailing the positive experiences that could be gained from having happier partners and better functioning relationships (Mikulincer & Shaver, 2007). However, Lemay and Dudley (2011) suggested that the partners of insecure individuals can regulate those individuals’ insecurity to protect the relationship. The following describes how we have conceptualized and studied, with a focus on conflict, the ways partners can buffer anxious and avoidant individuals in relationship-threatening contexts.
The Dyadic Regulation Model of Insecurity Buffering
Our research has been guided by the dyadic regulation model of insecurity buffering (see Fig. 1 for a schematic illustration), which considers how couples behave during attachment-relevant dyadic interactions (see also Overall & Simpson, 2013). As the top left of Figure 1 shows, stressful/threatening events activate the prototypic concerns of insecurely attached people (targets). The nondistressed partner (agent), however, may enact buffering behaviors to reduce (downregulate) the target’s distress and console him or her. Buffering behaviors can be enacted deliberately (consciously and intentionally) or automatically (nonconsciously or unintentionally) by agents and may include offering reassurance of continued love and support, accommodating the target’s wishes/needs, using “softer” influence tactics when trying to persuade the target during conflicts, and providing the right type of support the target needs to regulate his or her emotions more constructively.
Schematic illustration of the dyadic regulation model of insecurity buffering. The examples in this model focus on the ways in which partners can downregulate insecure reactions associated with attachment anxiety and avoidance during attachment-relevant behavioral interactions. The model can also be applied to other forms of insecurity (e.g., neuroticism, rejection sensitivity, low self-esteem), but the specific type of partner-buffering behavior must be responsive to the needs and motives of the target’s form of insecurity (see Lemay & Dudley, 2011, for a different example of how partners can regulate chronic relationship insecurities on a daily basis).
Buffering ought to be successful when the agent’s behavior is responsive to the particular concerns/needs associated with the target’s attachment orientation. Anxiously attached targets should benefit most from buffering behaviors that reassure them they are loved and supported, such as providing sufficient emotional support, attempting to fulfill their wishes and needs, or assuaging their relationship-relevant concerns. Avoidantly attached targets should benefit more from buffering behaviors that permit them to maintain their autonomy and independence, such as using softer influence tactics when trying to change their traits or opinions and providing instrumental forms of support designed to “fix” the problem and meet their needs while allowing them to retain personal control and remain self-reliant.
If the agent’s buffering works, insecure targets should report greater felt security during stressful/threatening interactions. This increase in felt security, in turn, should lead them to feel less distressed, manage their emotions better, and behave more constructively (see the middle of Fig. 1). If this pattern of stress/threat→responsive partner buffering→positive relationship perceptions and behaviors occurs repeatedly, targets should come to view themselves more positively and report greater overall relationship satisfaction, and agents should experience better relationship outcomes with their better adjusted insecure partners (see the right of Fig. 1).
Returning to Matthew and Helen, if Matthew (who is anxiously attached) feels threatened by a major disagreement that he and Helen are having, Helen may directly reassure Matthew of her unwavering love, recounting all of his wonderful traits and why she remains so steadfastly committed to their relationship. This buffering attempt should make Matthew feel more secure in the discussion, which should help him feel better, control his volatile emotions, and act more positively toward Helen. If this cycle continually reoccurs, Matthew should begin to automatically associate Helen with the dissipation of negative thoughts and feelings whenever he feels threatened, thereby strengthening the emotional bond between them (especially for Matthew; see Simpson, 2007). Over time, Matthew should begin to view himself as a valued partner, he should trust Helen more, and their relationship should improve.
Behavioral Dyadic Studies
We have conducted several behavioral observation studies with romantic couples to test components of our model. We have focused on conflict because this threatening context should activate the insecurities of both anxiously attached and avoidantly attached people.
Buffering anxiety during conflict
In some of our research, we have investigated how partners buffer the destructive reactions to conflict commonly displayed by anxiously attached individuals. For example, in one study, we videotaped married couples discussing important habits they wanted to change in each other, which elicits fears of rejection in anxious people (Tran & Simpson, 2009). We measured each partner’s emotional reactions and then had trained raters code each partner’s accommodation behaviors (e.g., inhibiting the urge to hurt or retaliate against the partner, trying to maintain the relationship by calming the partner and working to solve the problem). Anxious individuals felt more negative emotions and displayed less behaviorally rated accommodation during the discussions. However, the partners of anxious individuals displayed more accommodative behaviors if they themselves were more committed to the relationship, and these actions led anxious individuals to report more acceptance and positive emotions during their discussions. In other words, the immediate benefits of partner buffering were observed: Stronger partner commitment and more behavioral accommodation by the partner allayed the fears and improved the threat-based reactions of anxious individuals, thereby producing greater felt security and more constructive emotions and behaviors (see the middle of Fig. 1).
Buffering avoidance during conflict
We have also investigated the behaviors that buffer avoidantly attached individuals. In another study, we videotaped romantic couples discussing relationship problems identified by one partner (the agent) who wanted changes in the other partner (the target; Overall, Simpson, & Struthers, 2013). Given their need for autonomy, being targeted for change should be particularly threatening for avoidant people. As predicted, avoidant targets felt greater anger and displayed more coder-rated withdrawal during these discussions, which resulted in less successful problem resolution. Some partners, however, buffered avoidant defenses by softening their influence attempts, which involved being sensitive to the target’s autonomy needs, validating his or her viewpoint, and acknowledging his or her constructive efforts and good qualities. Avoidant targets whose partners displayed more softening exhibited less anger and withdrawal, and their discussions were more successful.
Simpson, Winterheld, Rholes, and Oriña (2007) had romantic partners complete the Adult Attachment Interview, which assesses attachment orientations to one’s parents. One week later, each couple was videotaped trying to resolve their most important relationship problem. At peak distress points during each discussion (the triggering event in Fig. 1), coders rated the extent to which (a) the less distressed partner (the agent) displayed emotional, instrumental, or physical caregiving behaviors; and (b) the more distressed partner (the target) was calmed by his or her partner’s caregiving attempts. Securely attached individuals were rated as more calmed when their partners gave them more emotional care (e.g., encouraging them to talk about their emotions/experiences with the problem, conveying unequivocal emotional support). However, because they manage stress by suppressing their emotions and limiting emotional closeness with partners, emotional caregiving should exacerbate distress in avoidant individuals. Accordingly, avoidant individuals were rated as more calmed when their partners delivered less emotionally imbued and more instrumental caregiving (e.g., giving concrete advice/suggestions for how to solve the problem, discussing the problem in an intellectual/rational manner). These findings confirm that, to be effective, partner-buffering behaviors must be tailored to the specific needs, concerns, and defenses of insecure targets.
Buffering insecurity after conflict
Finally, we have examined what individuals do to buffer their insecure partners in the aftermath of conflict discussions. In a longitudinal study by Salvatore, Kuo, Steele, Simpson, and Collins (2011), one partner in each couple had been studied since birth, so we had childhood attachment scores (rated in Ainsworth’s Strange Situation) for these individuals. Immediately after a videotaped discussion of a major relationship problem, each couple did a “cool-down” discussion task during which they talked about the most positive aspect of their relationship. We assessed how quickly and completely each partner “recovered” from the prior conflict discussion. Insecurely attached individuals (rated as insecure as children 20 years earlier) had more trouble recovering and were more likely to “reengage” the conflict during the cool-down task. However, when their partners displayed better conflict recovery, insecure individuals felt much more positive about the relationship, and these couples were more likely to still be dating 2 years later. Thus, as depicted on the right of Figure 1, partner buffering produced more beneficial longer-term outcomes, including stronger relationship stability across time.
Conclusions, Caveats, and Applications
These studies of romantic couples engaged in conflict highlight the critical role that partner buffering plays in protecting relationships that have insecurely attached partners. We have focused on conflict because it is threatening to both anxious and avoidant people, albeit for different reasons. However, partner buffering should also occur in other situations, such as when insecure targets need support. When partners (agents) provide more visible/direct support, anxious people tend to be happier and more optimistic about their relationships (Campbell, Simpson, Boldry, & Kashy, 2005). Conversely, because avoidant people prefer to manage stress by themselves, more indirect, less emotionally focused forms of support are required to circumvent their avoidant defenses (Simpson et al., 2007). Buffering behaviors are also important outside the context of threatening interactions. Lemay and Dudley (2011) found that partners exaggerate their affection when they perceive that targets feel insecure, which makes anxious individuals feel more accepted and secure. More frequent or satisfying sex can also improve the perceived emotional availability of partners (agents), which buffers insecure individuals (targets) from relationship dissatisfaction (Little, McNulty, & Russell, 2010).
Our research and the other studies described herein show that attachment insecurity does not spell doom for insecure people or their relationships; partners who enact appropriate behaviors can—and do—buffer insecurity. Certain components of our model need to be tested more fully, and additional partner behaviors not depicted in our model might also buffer insecurity. Although we focused on attachment insecurity, our model can also be applied to other forms of insecurity (e.g., neuroticism, rejection sensitivity, low self-esteem) known to produce emotion- and behavior-regulation difficulties in stressful or threatening situations. Partner-buffering behaviors will differ in effectiveness depending on how well and directly they address the underlying motives and reasons for a target’s regulation difficulties. However, even the “right kind” of buffering behaviors must be delivered skillfully; exaggerated or prolonged buffering attempts could be perceived as insincere by insecure targets, thereby failing to quell their distress (Lemay & Clark, 2008). When buffering attempts repeatedly fail, agents may burn out and stop trying to console insecure targets, which amplifies agents’ dissatisfaction (Lemay & Dudley, 2011). But when partner buffering successfully counteracts the worries and defenses of insecure people, persistent partner buffering may yield greater security across time. Partner buffering, therefore, may be a primary “agent of change” in making people more secure.
In conclusion, partner buffering is an important and understudied dyadic process that deserves more attention. Current studies have revealed how appropriate partner buffering can assuage negative reactions associated with attachment insecurity in relationship-threatening contexts and lead to better long-term outcomes for both insecure people and their partners. We hope that this framework will motivate researchers to identify dyadic processes that buffer other forms of insecurity and help therapists to stabilize and improve relationships that have insecure partners.
Recommended Reading
Overall, N. C., & Simpson, J. A. (2013). (See References). A general, comprehensive overview of the operation of partnerregulation processes in romantic relationships.
Overall, N. C., Simpson, J. A., & Struthers, H. (2013). (See References). An empirical example of how the use of “soft” influence tactics can buffer and regulate avoidantly attached partners.
Salvatore, J. E., Kuo, S. I., Steele, R. D., Simpson, J. A., & Collins, W. A. (2011). (See References). Based on data from the Minnesota Longitudinal Study of Risk and Adaptation, an empirical example of how individuals can facilitate faster emotional recovery from conflict in their insecurely attached partners.
Simpson, J. A., Winterheld, H. A., Rholes, W. S., & Oriña, M. M. (2007). (See References). An empirical example of how emotional and instrumental forms of caregiving can be used to reduce distress in anxiously and avoidantly attached partners.
Tran, S., & Simpson, J. A. (2009). (See References). An empirical example of how an individual’s commitment and accommodating behavior makes anxiously attached partners feel and behave more constructively during conflict discussions.
Footnotes
Author Contributions
Jeffry A. Simpson and Nickola C. Overall both contributed to the development of the ideas and the writing of all parts of this article.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.
Funding
This research was supported by grants from the National Institute of Mental Health to J. A. Simpson (R01-MH49599) and from the Royal Society of New Zealand Marsden Fund (UOA0811) to N. C. Overall.