Can Less Ever Be More? A Model of Emotion Regulation Repertoire of Social Support (ERROSS)

Structural Social Support

Considerable evidence suggests positive associations between structural social support (e.g., network size) and psychological and physical well-being (Cohen & Wills, 1985; Hansen et al., 2017; Holt-Lunstad et al., 2015; Lund et al., 2000). Individuals with more social ties report less distress and greater positive affect regardless of underlying stress levels (main effect hypothesis; Cohen & Wills, 1985). However, social support benefits can also be derived from small networks. Research on attachment figures and dyadic coping shows that even one or two close ties can be associated with improved affect regulation and health (e.g., Debrot et al., 2013; Horn et al., 2018). An analysis of online health-related social networking sites found more activity and support in smaller compared to larger networks (Hether et al., 2016). Thus, smaller networks seem to facilitate interaction among members, whereas larger networks may diffuse support responsibilities (Hether et al., 2016). Furthermore, larger networks are not always related to positive outcomes, as indicated by small or nonsignificant effects (Chu et al., 2010; Huxhold et al., 2013; Pressman et al., 2005). These findings suggest that larger networks are not necessarily associated with better outcomes than smaller networks. Instead, both large and small networks can be predictive of good outcomes, and additional variables may be needed to explain the relationship between social support and outcomes.

One possible explanation for the oftentimes overlooked positive effects of small networks on individuals’ health is that structural aspects of social networks don’t automatically reflect the array of support (e.g., quality, type) offered by network members. For example, a small network, although seemingly inadequate, could promote individuals’ well-being as long as individuals are satisfied with the support they receive. A large network, in contrast, may at first seem superior but can fall short of the expected benefits if network members are not reliable sources of support, offer the wrong support, or are not perceived as helpful. This notion is also evidenced by only moderate correlations between structural and functional aspects of social relationships (Coyle & Dugan, 2012). Thus, assessing only the structure of social networks provides little information about the type, quality, amount, and perceived adequacy of support and its implications for well-being.

Emotional Social Support

Emotional social support, or “the ability to turn to others for comfort and security during times of stress, leading the person to feel that he or she is cared for” (Cutrona & Russell, 1990, p. 99), appears to constitute a crucial facet of functional support. It is at the heart of many current theories of social support and relationships. For instance, simply the perception that a relationship partner is responsive to one’s goals, needs, and values is presumed to contribute to the couple’s intimacy and closeness (Reis et al., 2004). The key role of emotional support is also depicted in the relationship enhancement model of social support (Cutrona et al., 2005), in which interpersonal trust (i.e., the perception of one’s partner as kind, loving, and caring) represents a main mechanism in the link between social support and health. Indeed, emotional support predicts both personal and relational well-being (Gable et al., 2006; Gleason et al., 2008), and is linked to a number of positive outcomes including fewer sleep disturbances, better cognitive functioning, and reduced mortality risk (Bolger & Amarel, 2007; Lyyra & Heikkinen, 2006; Martire et al., 1998; Nordin et al., 2012; Seeman et al., 2001). This functional component of social relationships is suggested to operate through a stress-buffering mechanism (Cohen & Wills, 1985), with emotionally supportive relationships buffering the effects of stress by enhancing individuals’ coping abilities. Emotional social support frequently emerges as a better predictor of individuals’ health beyond structural and other functional aspects of networks (Berkman et al., 1992; Gleason et al., 2008).

Most often, emotional support is offered by few close network members rather than loose ties (Dunbar & Spoors, 1995; Hill & Dunbar, 2003; Stiller & Dunbar, 2007). The belongingness–bonding–binding continuum and other clusters of relationships suggest that social networks are comprised of different layers, including the most intimate connections (e.g., marital ties), close support connections (e.g., best friends, close relatives), and weaker relations (e.g., community members), with layers increasing in size but decreasing in relationship intensity (Dunbar & Spoors, 1995; Hill & Dunbar, 2003; Lin et al., 1999). Closer ties are relationships from whom one would seek advice, support, or help in times of emotional distress, whereas the distal layer characterizes loose social ties which provide a sense of belonginess but less emotional support (Dunbar & Spoors, 1995; Hill & Dunbar, 2003). The number of close ties that provide emotional support is frequently small and restricted to approximately two to 10 network members (Hill & Dunbar, 2003; Stiller & Dunbar, 2007). Even though many individuals seem to derive emotional support from a small circle of people, social networks are characterized by substantial interindividual variation, with some individuals relying on more and others relying on few social ties for emotional support (Dunbar & Spoors, 1995; Hill & Dunbar, 2003). Armstrong and Kammrath (2015), for instance, reported distinct associations between individuals’ enacted support-seeking strategies and their measured outcomes. Whereas participants perceived higher levels of support availability when seeking support from different network members (rather than more support from few people), higher self-esteem was reported when seeking support broadly and deeply (a lot of support sought from different network members).

To summarize, there is no clear evidence of the superiority of having many social ties versus a smaller number of relationships that nevertheless offer high levels of support. Structural aspects of networks are not necessarily a good indicator of the types of support offered by network members. The weak correlations between different social support components highlight the importance of distinguishing between structural and functional aspects of support, and the need to assess both in order to derive implications for mental health. Emotional support, perhaps the most important domain of social support, seems to be of particular value for individuals’ personal and relational well-being (Gleason et al., 2008), with great interindividual variability in how people utilize others for emotional support (Armstrong & Kammrath, 2015; Dunbar & Spoors, 1995).

Intrapersonal Emotion Regulation

Gross’s process model of emotion regulation has been highly influential in intrapersonal regulation-focused research (Gross, 1998). A recent extension of the original process model highlights how individuals identify opportunities to regulate emotions, initiate regulation, and how particular strategies are selected and implemented (Gross, 2015b). Bonanno and Burton (2013) proposed a regulatory flexibility model, emphasizing that the use of any one strategy is less important than the ability to vary one’s responding across different stressors that call for different strategies. The range of emotion regulation strategies that a person is able to use is called the emotion regulation repertoire. High repertoire flexibility is associated with better psychological functioning, such as decreased levels of distress or symptoms of posttraumatic stress disorder (Bonanno et al., 2004; Burton et al., 2012; Levy-Gigi et al., 2016). Due to the importance of repertoire in our model, we will come back to this concept in a later section.

Interpersonal Emotion Regulation

The present article defines IER as individuals’ goal-directed attempts to manage their own emotions through social partners and the emotion regulation strategies those partners afford. We propose that emotion regulation strategies afforded by relationship partners serve as IER (e.g., encouragement from a partner may reduce one’s own anger). This notion is supported by findings that people seek out others to share their emotions (Gable et al., 2004; Rimé, 2009), and selectively turn to different relationship partners for different emotions (Cheung et al., 2015).

Other definitions of IER distinguish between efforts to modify one’s own (intrinsic), another person’s (extrinsic; Zaki & Williams, 2013), or each other’s emotional states (dyadic; (Butler, 2015; Butler & Randall, 2013), or consider the extent to which IER requires feedback from others (i.e., response-dependent vs. independent; Zaki & Williams, 2013) and is intentional or deliberate in nature (Koole, 2009; Niven, 2017). Since the emotion regulation repertoire of social support (ERROSS) model introduced in this article focuses on the target of regulation (i.e., the person receiving emotion regulation strategies) rather than the regulator’s perspective (i.e., the network member offering strategies) or a dyad perspective, our definition of IER does not include these additional considerations and we direct the interested reader to Dixon-Gordon et al. (2015); Niven (2017); and Zaki and Williams (2013).

IER can take on various forms from humor (Horn et al., 2018), to encouragement (Niven et al., 2015), to drawing on a friend’s reappraisal skills (Horn & Maercker, 2016). It can be distinguished from related concepts such as emotional support and dyadic coping, as well as from intrapersonal emotion regulation (Williams et al., 2018; see previous description). Unlike emotional support, IER is goal-directed and targeted towards modifying emotional experiences, even outside the context of stress. Emotional support, in contrast, is a more general concept than time-limited, emotional-event-focused IER, and is offered in response to stress (e.g., during the time period following a diagnosis of chronic illness or unemployment). Emotional support can be employed in response to both emotional and nonemotional goals (Cutrona & Russell, 1990). For example, emotional support may have the goal of having a person persist in chemotherapeutic treatment, but not necessarily to make them less afraid. In addition, emotional support is one form through which people use others to regulate their emotions (e.g., providing comfort). Dyadic coping describes how two individuals in a close relationship respond to stressors jointly (Bodenmann, 1997), whereas IER can be viewed from various perspectives (e.g., see intrinsic vs. extrinsic IER; Zaki & Williams, 2013), not necessarily involving both partners and their oscillating patterns of regulation.

IER is common and can help individuals effectively manage both positive and negative emotions. Successful IER is associated with adaptive adjustment, such as increased momentary affect, psychological well-being, and heightened relationship quality (Horn & Maercker, 2016; Horn et al., 2018; Niven et al., 2015; Niven, Holman, & Totterdell, 2012). For example, participants’ neural response to threatening electric shocks was attenuated when they held the hand of either a stranger or their spouse, though this hand-holding had a greater effect for couples with higher marital quality (Coan et al., 2006).

Which and when strategies promote successful IER is still unclear (Niven et al., 2015; Niven et al., 2009). When it comes to regulating one’s own feelings, certain strategies (e.g., reappraisal) are indeed considered more effective than others (Gross & John, 2003), but only to the extent that these strategies meet contextual demands across time (Bonanno & Burton, 2013; Webb et al., 2012). For example, a meta-analysis by Webb et al. (2012) found unique associations in the effectiveness of intrapersonal emotion regulation both within and between strategies. While cognitive change processes were more effective than attentional deployment and response modulation processes in regulating emotional experiences, different subtypes within these emotion regulation processes were associated with distinct outcomes. If this finding extends beyond the intrapersonal context, one could expect that specific IER strategies are less important than the availability of a pool of different IER strategies. If your friends and family members cover a variety of different strategies (e.g., reappraisal, distraction, problem-solving, and encouragement), there is a greater likelihood that one of these strategies will cover your regulatory need at a given moment. In times of need (e.g., when arguing with your partner), one could then choose from the array of IER strategies offered by network members and flexibly draw on the strategies that seem most effective based on contextual demands, rather than relying exclusively on one strategy across different situations. Conversely, if an IER attempt fails, a pool of diverse IER strategies can provide additional opportunities for IER. For instance, if your friend’s attempt to put your ex-partner in a bad light is not effectively helping you manage your emotions, you could make use of other IER strategies from the repertoire (e.g., distraction, reassurance, reappraisal). Building on advancements in the field of emotion self-regulation suggesting that flexible use of diverse intrapersonal strategies is associated with benefits (see regulatory flexibility; Bonanno & Burton, 2013; Levy-Gigi et al., 2016), the availability of different IER strategies should promote context-sensitive IER responding and, in turn, increase adaptation.

Benefits of a repertoire of diverse emotion regulation relationships have been recently studied. Evidence suggests that people indeed selectively turn to specific individuals who they consider effective at regulating a particular emotion—termed emotionships (Cheung et al., 2015). In a study by Cheung et al. (2015), individuals were able to name discrete relationship partners for specific emotions (e.g., relationship partner effective at cheering up sadness vs. soothing anxiety) and drew closer to these nominated individuals when the specific emotion was activated. Interestingly, greater well-being was reported when IER efforts were diversified across specialized social ties. Similarly, among college students, empathic individuals were sought out for trust, serving as stress buffers, and had central roles in smaller networks, whereas students high in well-being attracted fun and excitement, fostering positive experiences, and occupied central roles in broader networks (Morelli et al., 2015). The evidence outlined before suggests that individuals have a sense of whom to turn to for IER, and that benefits may be maximized when IER is diversified (Cheung et al., 2015) and when individuals can draw on different regulation strategies (Bonanno & Burton, 2013). In the study by Cheung et al. (2015), participants identified people they would approach for particular emotional experiences (e.g., when experiencing sadness vs. anger). Distinct, however, is the question whether emotional experiences correspond to regulation strategies afforded by network members. It seems unlikely that an individual would always turn to the same friend when experiencing anger. For instance, when the anger results from a rejected manuscript, a person might be more inclined to turn to an academic colleague to discuss the rejection, but to another network member when anger is caused by relationship problems and distraction is sought. Who the regulator is, how he or she responds (i.e., IER strategy), and to what extent these IER strategies are concentrated on few versus many relationship partners may have stronger implications with regard to network usage and well-being.

In sum, existing frameworks and empirical findings highlight the importance of others in helping people regulate their emotions. Whereas some individuals draw on a large network, others rely on the same few people. Research on intrapersonal emotion regulation processes suggests that a pool of different strategies promotes context-sensitive responding and health benefits. To what extent emotion regulation repertoire can be extended to encompass interpersonal aspects and why both small and large networks are sometimes associated with well-being are the core questions that drove the development of our emotion regulation repertoire of social support (ERROSS) model. ERROSS proposes that effective social support operates through various emotion regulation strategies provided by network member(s), which are then available to the support seeker. We posit that a diverse network of IER strategies offered by others can provide people with a broader pool of resources to draw from, thereby allowing them to capitalize on others’ strengths in providing regulation. Further, we hypothesize that people will have greater well-being if an individual’s network offers a variety of loved ones, or a variety of different IER strategies. Of course, the best outcomes would follow for people who have both. This conceptualization of effective social support as a combination of structural support and emotion regulation strategies constitutes a unique approach to assessing network utilization in IER. Of course, we do recognize that human interactions are complex for a number of reasons. For instance, individuals need to be aware of their need for IER and able to communicate it for relationship partners to be able to respond. Even if someone seeks help, regulators may not be able to decode the help-seeking attempt, they may misread the intent, or the supposedly helpful emotion regulation strategies may not be in tune with the expresser’s goals (e.g., if they simply want to vent and the regulator provides advice rather than just listening). There are costs of mismatches between needs and IER attempts, contextual constraints, individual differences in delivery of IER, and iterative mutual influences within the dyad. However, for this article, we present a simplified model. As we describe ERROSS, we will first introduce a basic, computational model before depicting the model in a more nuanced, complex version. This computational model is an intentional oversimplification which introduces its core components and main assumptions. Although the model is simplified, we believe that researchers can still benefit from this model, as it allows for an easy generation of research hypotheses. In a similar vein, we introduce the basic model in terms of global perceptions first and consider the role of actual social exchanges in a later section.

Building Blocks: Network Size, Emotion Regulation Strategies, and Repertoire

Within ERROSS, three main variables are operationalized: network size, emotion regulation strategies, and emotion regulation repertoire.

Network size

Network size is defined as the number of close others with whom individuals are in regular contact (see Social Network Index; Cohen et al., 1997). Building on research findings which suggest that emotional support is typically provided by close social ties (Dunbar & Spoors, 1995; Hill & Dunbar, 2003; Stiller & Dunbar, 2007), our definition of network members does not include loose ties. In Figure 1, network size is graphically depicted as the number of columns with at least one black and white checkered cell. In this example, Figures 1a and 1c have a network size of two, 1b a size of 10, and 1d a network size of eight. Table 1 provides an overview of arithmetic operationalization and graphic depiction of the various model components as well as the resulting model component estimates for the four hypothetical individuals presented in Figure 1.

OPEN IN VIEWER

Figure 1.

Exemplars of interpersonal emotion regulation repertoires from four hypothetical individuals, including network size and emotion regulation strategies provided.

OPEN IN VIEWER

Table 1.

Components of ERROSS, including their arithmetic operationalization, graphic depiction, and their estimates for four hypothetical individuals from Figure 1.

Component of ERROSS	Arithmetic operationalization	Graphic depiction in Figure 1	Representation in Figure 1
			Figure 1a	Figure 1b	Figure 1c	Figure 1d
Network size	Count: Number of distinct network members	Columns with at least one cell checkered	2	10	2	8
Emotion regulation strategies	Count: Number of distinct emotion regulation strategies	Rows with at least one cell checkered	2	2	10	10
Emotion regulation repertoire	Count: Number of checkered cells (total number of nondistinct emotion regulation strategies available)	Checkered cells	2	12	12	24

Note. ERROSS = emotion regulation repertoire of social support model.

Main Assumptions

Building on existing research from the social support and emotion regulation literature cited before, ERROSS postulates specific hypotheses regarding the mental health associations of network size, emotion regulation strategies, and repertoire. Whereas assumptions regarding network size and emotion regulation strategies will only be discussed briefly, we will focus on repertoire due to its novelty and since this variable is at the heart of our model.

At its core, ERROSS suggests that effective socioemotional support operates through the availability of various emotion regulation strategies provided by network member(s), resulting in a repertoire of strategies the support seeker can draw on. Repertoire, the conglomerate of total network size and total number of emotion regulation strategies afforded, is assumed to predict good mental health, such that individuals with a larger repertoire should report higher well-being compared to individuals with a smaller repertoire (main effect). This assumption follows from Bonanno and Burton’s (2013) regulatory flexibility model which suggests that the use of any one strategy is less important than the ability to vary one’s response across different stressors that call for different strategies (Burton et al., 2012; Levy-Gigi et al., 2016). In the interpersonal context, access to various emotion regulation strategies across different network members should promote context-sensitive responding by allowing individuals to have choices for the best-fitting strategies through multiple relationship members.

Furthermore, we hypothesize that people will have greater well-being if an individual’s network offers a variety of loved ones, or a variety of different IER strategies (main effects). Building on the general positive relationship between network size and health (Hansen et al., 2017; Holt-Lunstad et al., 2015), larger networks are not only positively associated with perceptions of support, but they also increase the likelihood of actual support. Similarly, different strategies provide an array of emotion regulation strategies to draw from, and enhance the likelihood that at least one of the strategies will meet the regulatory need, which is adaptive for individuals’ adjustment to life stressors (Orcutt et al., 2014). Even though this relationship has only been shown for intrapersonal emotion regulation, it is plausible that the same is true in an interpersonal context. To summarize, ERROSS proposes that network size, emotion regulation strategies, and repertoire function as predictors of individuals’ well-being via main effects, with greater benefits posited for those with access to larger networks, different strategies, and greater repertoire.

So far, ERROSS has been described primarily as the global perception of social networks and the IER provided by their network members. The model, however, can also be applied to individuals’ actual social exchanges. We posit that the ERROSS model’s main assumptions would hold for both perceptions and actual social exchanges, but that pathways may differ for the proposed relationships. As such, perceptions of a large network, different emotion regulation strategies, and a large repertoire may exert their effect through increasing overall levels of perceived support. On the other hand, actual access to emotionally supportive others may impact well-being through increasing an individual’s coping resources, such as promoting intrapersonal emotion regulation and context-sensitive responding, or decreasing efforts associated with intrapersonal emotion regulation. In addition, larger effects are expected for individuals’ perceptions versus actual social exchanges, given past research that has consistently documented positive associations between perceived support and well-being (Kawachi & Berkman, 2001; McDowell & Serovich, 2007; Reinhardt et al., 2006), whereas actual support interactions frequently fail to yield positive outcomes (Bolger et al., 2000; McDowell & Serovich, 2007; Shrout et al., 2006).

ERROSS: Individual Differences and Contextual Factors

The extent to which emotion regulation afforded by others is associated with mental health benefits likely depends on complex situational and individual features. Although our model is an attempt to focus on the predictive power of emotion regulation repertoire, in the following, we highlight selected situational factors and individual differences that may moderate or mediate the mental health associations of the ERROSS model’s variables.

For example, provision of emotion regulation may come at a relational cost, varying by situation and the person providing the regulation strategies. Network members may not always be willing to provide IER. Drawing on the same few people for IER (i.e., few columns and many rows in Figure 1) may exhaust the few network members, which we refer to as overburdening. Network members may also not want to provide IER because of unmet expectations of reciprocity, whereby they feel that they have provided more than they have received in return. In this sense, some level of redundancy of network members or IER strategies in one’s repertoire may prevent overburdening and buffer against temporal unavailability of network members. For example, Individual 1b has perfect redundancy of IER strategies (i.e., each strategy is offered by more than one network member), whereas 1a and 1c have less redundancy built in their repertoire and thus a higher chance of overburdening their relationship partners. Preliminary evidence by Cheung et al. (2015) suggests that individuals’ well-being is maximized if they diversify their emotion regulation needs across different relationship partners rather than concentrating all needs on the same few network members.

On the other hand, someone who has a wider network may accumulate more costs, suffering from what we call overhead. Relying on many different social network members requires more time, possibly more shame from imposing on multiple people, and more obligations for reciprocal support, among other costs. How individuals optimally arrange and utilize their social networks for IER, and whether overhead and overburdening play a role is currently unknown, but testable in the ERROSS model. Empirical tests of the ERROSS model may be able to mathematically assess tipping points for overburdening and overhead, and include these as predictors of mental health outcomes as well.

Thus, similar to feedback processes described in existing models of emotion regulation (Bonanno & Burton, 2013; Gross, 2015b), the effectiveness of network members and strategies in regulating affect will vary. The person in need of emotion regulation must monitor whether additional efforts are necessary. These feedback processes, in turn, will inform future selection and application of network members and strategies. Failures, or costs, of IER could appear at any stage of the process. For example, people could make suboptimal choices of interaction partners, miscommunicate their emotional needs, or overburden network members. We must also consider the possibility that greater use of IER might indicate ineffectiveness. Seeking out more IER from a larger number of relationship partners might suggest that the initial IER attempt was not successful. By the same logic, low usage could reflect a highly effective support (or other interpersonal) process, whereby the person’s emotion regulation needs are met and they are able to adjust or feel better without needing to seek out more IER strategies and/or network members.

In an example of moderation, attachment style and use of IER resources may also impact perceptions of the supportiveness of one’s network as well as actual exchanges. Securely (vs. insecurely) attached individuals may not only report access to a larger repertoire, but these differences in perception may also translate into their actual support-seeking behavior and, in turn, affect their mental health.

Although access to a large repertoire should promote context-sensitive responding by allowing individuals to flexibly choose among the best-fitting strategies and relationship members, this requires not only the ability to identify and discriminate which network members are effective (vs. ineffective) at providing certain emotion regulation strategies, but also to selectively seek out network members. Research suggests that individuals differ on their tendency to rely on and their perceived efficacy of IER (Williams et al., 2018). Preliminary evidence suggests that individuals have an awareness of whom to turn to when experiencing specific emotions, and that people turn to these relationship partners when such an emotion is elicited (Cheung et al., 2015). If individuals lack such an awareness, a large repertoire should not be more beneficial than a small or medium-sized repertoire. Fortunately, perceived efficacy can be measured, and could be used as a moderating variable with empirical data.

Beyond individual differences, situational and contextual factors may also affect the ERROSS model. For instance, different sociodemographic and cultural groups may use and benefit from network size, emotion regulation strategies, and repertoire differently. According to the transactional model of stress and coping (Lazarus & Folkman, 1984), when the need for emotion regulation is high, it may be especially beneficial to rely on others for regulatory benefits. Thus, we present the example of bereavement to further explore the ERROSS model.

Emotion Regulation and Social Support During Bereavement

Stressful situations such as a romantic break up, job loss, or the death of a loved one commonly provoke a range of intense emotions. One unique aspect of the death of a loved one is its irreversibility, reducing options for changing the stressor situation. Therefore, bereaved individuals need to regulate their heightened emotional intensity and reactivity (Horowitz et al., 1997; Shear et al., 2011). During this time, emotion regulation facilitates the critical task of adaptation: integrating the loss and adjusting to life without the deceased. However, not all bereaved individuals have the self-regulatory resources to manage this emotional turmoil. For example, research suggests greater adjustment problems for survivors who experience more difficulties regulating their emotions (Diminich & Bonanno, 2014; Gupta & Bonanno, 2011).

Bereavement is an example of a situation in which the ERROSS model is particularly useful. When emotional intensity is high and sources of self-regulation are diminished, social ties are particularly important. Research suggests that the lack of social support constitutes a risk factor in the development of clinical conditions such as prolonged grief disorder or complicated grief, and adjustment problems in general (Burke et al., 2010; van der Houwen et al., 2010).

Cognitive stress theory and other theories of social support (Cohen, 1988; Cohen & Wills, 1985; Lazarus & Folkman, 1984) suggest that living loved ones can provide compensatory coping resources, even given the difficulties that may result from losing someone close like a spouse (i.e., an attachment figure). In fact, treatment for complicated grief targets interpersonal functioning by holding a conjoint session with a close, supportive person from the viewpoint that responsive, empathic, and trustworthy relationships can promote emotion regulation and adaptation to the loss (Shear, 2010, 2012).

The Role of Network Size, Emotion Regulation Strategies, and Repertoire During Bereavement

The ERROSS model expands on these existing perspectives by proposing that close relationships can help survivors manage the heightened emotional reactivity and potential regulatory challenges associated with grief if they provide access to different regulatory strategies. In turn, a lack of social ties may contribute to the development and maintenance of disordered grief because of the absence of others who could help regulate the emotions of the bereaved. In terms of the ERROSS model’s components, we expect the following associations for individuals’ adjustment to loss.