This is an open access article distributed under the terms of the Creative Commons
Attribution License (In the United States, approximately nine million people practice some degree of consensual non-monogamy, often polyamory (Sheff, 2014a). All individuals and relationships experience stressors. Polyamorous relationships experience the same potential sources of stress as monogamous relationships, managing illness, family stress, financial concerns, and relational tensions (Sheff, 2014b). However, polyamorous relationships must also legitimize their relationships to others, manage experiences of discrimination, and sometimes maintain multiple romantic relationships (Dixon, 2016). In addition, polyamorous people sometimes experience jealousy across multiple relationships that must be managed and maintained (Rubinsky, 2018, 2019). Relationships with close others can either facilitate productive coping, or intensify the harmful effects of stress (Afifi, 2018). Thus, relational contexts are uniquely positioned to buffer against or exacerbate the individual and relational health consequences of stress. How individuals manage stressful experiences is important because stress results in physical, emotional, and psychological health consequences (Afifi, 2018). The enactment of relational maintenance behaviors is crucial in maintaining satisfying polyamorous relationships and coping with experiences of stress, for which polyamorous individuals may experience in relation to their relationships and external relational perceptions (Dixon, 2016). The present study employs the theory of resilience and relational load (TRRL; Afifi, Merrill, & Davis, 2016) to explore the effects of repeated maintenance and investment behaviors on identity-laden stressors in polyamorous relationships. Exploring how daily maintenance behaviors moderate the effects of identity-laden stressors on relational satisfaction and resilience both expands opportunities for understanding productive relational behaviors in an understudied relational context, and tests and expands the emerging TRRL in polyamorous relationships.
Maintaining Resiliency in Multiple Partner Relationships
Relationships require communicative work to maintain. Relational maintenance can refer to a variety of behaviors used to keep a relationship in existence, at a desired level of involvement, intimacy, or satisfaction, or to prevent or repair relational problems (Dindia & Canary, 1993). Literature on relational maintenance behaviors generally categorizes these behaviors as either prosocial, which includes positivity, openness, assurances, networks, romance, affection, supportiveness, and sharing tasks, or antisocial, which includes argument, insolence, lack of self-disclosure, jealousy induction, infidelity, spying, destructive conflict, and allowing control (Canary & Stafford, 1992; Dainton & Gross, 2008).
The present study focuses on the seven dimensions of routine and strategic prosocial maintenance proposed by Stafford, Dainton, and Haas (2000). Positivity describes a presentation as cheerful and optimistic (Canary, Stafford, Hause, & Wallace, 1993). Openness describes direct discussion and disclosure between partners (Canary et al., 1993), especially about topics related to feelings for each other and the relationship (Stafford et al., 2000). Social networks refer to the use of common associations between partners to keep the relationship progressing (Canary et al., 1993), and the reliance on support and love by family and friends (Stafford et al., 2000). Sharing tasks describes fulfilling of chores and responsibilities within the relationship (Canary et al., 1993), and performing jointly faced tasks (Stafford et al., 2000). Assurance describes implicit or explicit reassurance about the future of the relationship (Stafford et al., 2000). Conflict management refers to integrative behaviors during conflict, like apologizing or admitting wrong-doing (Stafford et al., 2000). Finally, advice describes sharing opinions and advice with a partner about their problems (Stafford et al., 2000).
In addition, individuals in relationships also engage antisocial maintenance behaviors (Dainton & Gross, 2008). Jealousy induction describes intentional efforts to arouse feelings of jealousy in a partner (Dainton & Berkoski, 2013; Dainton & Gross, 2008). Avoidance describes avoiding particular topics in conversation as well as actively avoiding the partner (Dainton & Gross, 2008). Spying refers to surveillance attempts such as checking the partner’s phone. Destructive conflict describes intentionally starting arguments or engaging in controlling behaviors, as well as other unproductive conflict management styles (Dainton & Berkoski, 2013). Finally, allowing control describes behaviors like cancelling plans or events with family and friends, or failure to engage in activities the individual previously enjoyed, either to spend time with the partner or because a partner does not prefer those activities (Dainton & Gross, 2008). This also includes allowing the partner to make decisions or take control of the individual (Dainton & Berkoski, 2013). Negative maintenance behaviors also include infidelity, like flirting or having sex with other people to prevent boredom (Dainton & Berkoski, 2013; Dainton & Gross, 2008). However, because exploration or engagement outside of a primary relationship is often allowed within the polyamorous community (Dixon, 2016; Sheff, 2014b), infidelity is not considered as a negative maintenance strategy in the present study.
Lower quality relationships are more likely to engage in negative maintenance behaviors (Goodboy & Myers, 2010), but both positive and negative maintenance behaviors may be necessary to keep a relationship in existence (Goodboy & Myers, 2010). Prosocial maintenance behaviors consistently predict love, liking, satisfaction, and commitment (Dainton, 2000). Negative maintenance behaviors are negatively related to relational satisfaction (Dainton & Gross, 2008). Although many studies have explored the impact of prosocial and antisocial maintenance behaviors in the context of romantic relationships (Billedo, Kerkhof, & Finkenauer, 2015; Dainton, Goodboy, Borzea, & Goldman, 2017), research on polyamorous relationships have not engaged this specific construct, despite multiple studies alluding to the importance and influence of relational maintenance across multiple relationships (Rubinsky, 2018; Wosick-Correa, 2010).
Theory of Resilience and Relational Load
In general, relationships that engage in more positive maintenance behaviors may be more resilient (Afifi et al., 2016). Resilience is often conceptualized as the process of reintegrating from life’s disruptions by focusing on meaning-making through everyday messages and stories (Lucas & Buzzanell, 2012; Richardson, 2002). Communication is central to the resilience process (Lucas & Buzzanell, 2012). TRRL defines resilience as the ability to adapt positively when confronted with stress or adversity (Afifi, 2018; Afifi et al., 2016). According to TRRL, daily relational maintenance and communication behaviors allow relational partners to develop resiliency and thrive in their relationships. When people validate their relational partners in an on-going, prosocial manner, they accumulative positive emotional reserves that may safeguard their relationship from future stresses and challenges (Afifi et al., 2016). Further, TRRL argues that relational partners and families with a more communal orientation toward life and life’s stressors will invest more in their relationships and build increased emotional capital through repetitive, prosocial relational maintenance (Afifi et al., 2016). One reason for this, according to TRRL, is because individuals with more positive emotional reserves in their relationship are more likely to appraise stressful relational situations from a broader perspective and engage in uplifting communication; this uplifting communication helps to preserve the invested-in relationship (Afifi et al., 2016). These secure appraisals foster resilience and growth and minimize the experience of stress (Afifi et al., 2016).
However, in relationships in which the investments or standards for investments are unmet, individuals, according to TRRL, are more likely to engage in threatening appraisals and experience conflict during times of relational stress (Afifi et al., 2016). Threatening appraisals deplete cognitive, emotional, and relational resources, and rather than buffer against stress, they exacerbate feelings of stress (Afifi et al., 2016). TRRL introduces the concept of relational load, which occurs when long-term depletion wears away at the relationship. Relational load and resource depletion result in increased susceptibility to mental, physical, and relational health problems, and may result in less relational investment and a smaller likelihood of perceiving a communal orientation toward stress (Afifi et al., 2016).
TRRL is an emerging theory, but thus far has been applied to couples’ communication about financial uncertainty (Afifi et al., 2018). In addition, implicitly, couple identity gaps and conflict management within romantic relationships influence the experience stress (Merrill & Afifi, 2017). However, TRRL is still being applied across different relational types. The present study contends that polyamorous relationships are an important context within which to explore the ways that relational maintenance and investment can facilitate or impede positive relational outcomes because of the identity-laden nature of relational stressors in polyamorous relationships.
More Than Two: Polyamory and TRRL
Polyamory is a relational identity, sometimes described as responsible non-monogamy (Klesse, 2006), that allows for multiple partnerships or exploration of multiple romantic or sexual partnerships (Dixon, 2016). Ethical or consensual non-monogamy, like polyamory, involves awareness of the ability to explore or pursue other romantic and sexual bonds, and communication and negotiation about what that will look like in a given relational context (Klesse, 2011).
The polyamorous community promotes prosocial relational maintenance behaviors (Wosick-Correa, 2010). Relational maintenance scholarship is heavily rooted in concepts of equity, exchange, and relational investment (Dainton, 2000; Dainton et al., 2017), which have received attention from members of the polyamorous community. For instance, the polyamorous community shares relational advice and tips for managing multiple-partner relationships (Sheff, 2014b). Wosick-Correa (2010) notes a number of prosocial relational maintenance strategies either implicitly or explicitly emphasized by people who either identify with or practice polyamory. For instance, individuals who are polyamorous benefit from expanded social networks that could distribute emotional, physical, and even sometimes financial burdens (Conley & Moors, 2014; Mitchell et al., 2014; Wosick-Correa, 2010).
In addition, the polyamorous community discusses sharing tasks among relationships, noting how to avoid over-burdening a singular relationship (Conley & Moors, 2014). For example, new relationship energy (NRE) describes the familiar experience of excitement and warmth associated with a new relationship (Sheff, 2014b). Without undermining the pleasant feelings associated with this experience, people who are polyamorous communally advise in popular press books, websites, and meet-up groups, how to avoid letting one relationship’s NRE over-burden another relationship with more stressful life tasks (Sheff, 2014b). For instance, polyamorous individuals may describe how an individual with a new partner should be careful not to spend all of their time with a more established relational partner discussing financial or household tasks. Thus, the polyamorous community’s orientation toward emphasizing positive relational maintenance strategies suggests the potential for developing emotional reserves that can be relied upon during times of stress.
In addition, polyamorous relationships must manage stressors that may be unique compared to monogamous relationships. Considered a communicative identity, polyamory is counter to traditional notions of coupledom (Dixon, 2016). Further, managing jealousy has the potential to prompt relational stresses within polyamorous dynamics (Rubinsky, 2018, 2019). Thus, in addition to traditional relational stressors that occur in any relationship, polyamorous individuals often need to defend and legitimize their relationships to those outside the relationship (Dixon, 2016). In a relationship, stress affects not just the individuals but the relationship itself, which can buffer against or exacerbate feelings of stress (Afifi, 2018). TRRL argues that individuals in a relationship with a more unified approach toward a relational, individual, or life stressors, will be more likely to invest in relationships by maintaining them, which results in more positive outcomes (Afifi, 2018; Afifi et al., 2016).
Resilience and Relational Identity
Individuals invest in and identify with their relationships. As a result, they may experience stressors as a discrepancy between aspects of their identity (Merrill & Afifi, 2017). Identity gaps are dialectical tensions between layers of identity described in the communication theory of identity (CTI; Hecht, 1993). CTI conceptualizes an individual’s identity as occurring in four interconnected frames: the personal, enacted, relational, and communal (Hecht, 1993; Jung & Hecht, 2004, 2008). Personal identity refers to an individual’s sense of self. Enacted identity is the identity that exists in communication, expression, performances, or messages. Relational identity occurs at several levels, and includes ascriptions of identities assigned by others, identifying with a relational unit (e.g., couple, triad), and identifying in terms of relational memberships (e.g., daughter, partner). Communal layer refers to identifying with larger social groups (Hecht, 1993).
Discrepancies or tensions among these layers result in identity gaps. Merrill and Afifi (2017) conceptualized couple identity gaps to refer to the dyadic assessment of relational identity discrepancies. Couple identity gaps help explain how couples vary in the relational management of stress (Merrill & Afifi, 2017). However, since the present study uses individual-level measurement, it considers the more traditional personal-enacted and personal-relational identity gaps as cognitive, affective, and behavioral manifestations of identity-related relational stress at the individual level. Personal-enacted identity gaps describe situations when one’s self-concept is in tension to some extent with the self in communication (Jung & Hecht, 2004, 2008). Personal-relational identity gaps refer to experiences of cognitive, affective, or behavioral discrepancies between an individual’s sense of self or self-concept, and their ascribed relational identity (Jung & Hecht, 2004, 2008). Both identity gaps are negatively associated with relational satisfaction (Jung & Hecht, 2004, 2008; Jung, Hecht, & Wadsworth, 2007; Kam & Hecht, 2009), including in polyamorous relationships (Rubinsky, 2019).
The present study, framed through TRRL, contends that repeated use of prosocial maintenance behaviors demonstrates relational investment, and thus minimizes the experience of identity gaps, and bolsters relational satisfaction and resilience. Alternatively, negative maintenance behaviors may heighten the negative effects of identity gaps on relational satisfaction and resilience. This is important to address because despite the increasing prominence of polyamorous relationships and the implicit notion of the impact of relational maintenance behaviors on identity-laden aspects of relational stressors, little empirical work has explored these phenomena. In addition, exploring the potential of maintenance behaviors to reduce the impact of identity-laden stressors may provide practical information for those in polyamorous relationships or counselors who work with polyamorous clients. Thus, the following overarching hypothesis is proposed: Relational maintenance behaviors moderate the effects of identity gaps on relational satisfaction and resilience in polyamorous relationships. Specifically, this study hypothesizes:
In polyamorous relationships: H1: prosocial relational maintenance behaviors moderate the effects of (a) personal-enacted identity gaps and (b) personal-relational identity gaps on relational satisfaction. H2: Prosocial relational maintenance behaviors moderate the effects of (a) personal-enacted identity gaps and (b) personal-relational identity gaps on resilience. H3: Antisocial relational maintenance behaviors moderate the effects (a) personal-enacted identity gaps and (b) personal-relational identity gaps on relational satisfaction. H4: Antisocial relational maintenance behaviors moderate the effects of (a) personal-enacted identity gaps and (b) personal-relational identity gaps on resilience.
Testing this hypothesis will further advance TRRL as a framework for understanding how relationships cope with stress. In addition, it meets Afifi and colleagues’ (2016) call to explore TRRL in a diversity of relational types.
Method
Participants
Participants (n = 157) identified as being in at least one polyamorous relationship at the time of participation. Ages range from 18-81 (M = 38.84, SD = 13.31). Most participants identified as being assigned female at birth (n = 113, 71.97%), with the rest identifying as assigned male at birth, and two identifying as outside of those categories. In terms of gender identity, participants were primarily identified as cisgender women (n = 83, 54.25%), cisgender men (n = 35, 22.88%), genderqueer (n = 13, 8.5%), and genderfluid (n = 7, 4.58%), with the remainder identifying otherwise under the transgender or non-binary umbrellas. Regarding sexual identity, participants were heterosexual (n = 48, 30.57%), bisexual (n = 43, 21.02%), pansexual (n = 33, 1.91%), gay or lesbian (n = 3, 1.91%), and the rest broadly under the LGBTQ umbrella. Finally, participants were primarily white (n = 144, 90.57%) and two or more races (n = 12, 7.55%).
Procedures and Data Analysis
As part of a larger study on polyamorous relationship communication, participants consisted of a volunteer, purposive sample who completed a survey hosted on Qualtrics.com. Recruitment occurred on the researcher’s social media pages, polyamorous groups on Reddit and Tumblr, and via a polyamorous research listserv. Participants were directed first to an online informed consent form. If they agreed to participate, they were directed to a Qualtrics questionnaire. The questionnaire contained several parts. The measures described in this article are identified in the instrumentation section. To address the hypotheses, a series of moderation analyses with each maintenance behavior considered as a separate moderator were conducted with the PROCESS Macro in IBM SPSS (Hayes, 2012).
Instrumentation
Prosocial Maintenance
The strategic maintenance scale (Stafford et al., 2000) measures routine and strategic relational maintenance and involves seven factors: assurances, openness, conflict management, shared tasks, positivity, advice, and social networks. Minor adaptations were made to allow for multiple partners, i.e., partner to partner(s). Responses range from one = almost never to seven = almost always. Example questions include “I say ‘I love you’ to my partner(s)” and “I show my love for my partner(s).” Assurances contained eight items (Cronbach’s α = .89, M = 6.17, SD = 0.85). Openness contained seven items (Cronbach’s α = .88, M = 6.18, SD = 0.85). Conflict management contained five items (Cronbach’s α = .87, M = 6.16, SD = 0.78). Shared tasks contained five items (Cronbach’s α = .89, M = 5.85, SD = 1.02). Positivity contained two items (Cronbach’s α = .84, M = 5.53, SD = 1.13). Advice contained two items (Cronbach’s α = .81, M = 4.82, SD = 1.36). Social networks contained two items (Cronbach’s α = .75, M = 5.25, SD = 1.21).
Negative Maintenance
The negative maintenance scale (Dainton & Gross, 2008) measures six factors of negative relational maintenance including jealousy induction, avoidance, spying, infidelity, destructive conflict, and allowing control. An example question includes “I flirt with others to make my partner(s) jealous.” Infidelity was not included. Responses range from one = almost never to seven = almost always. Jealousy induction contained two items (Cronbach α = .92, M = 1.21, SD = 0.63). Avoidance contained four items (Cronbach α = .77, M = 2.66, SD = 1.24). Spying contained two items (Cronbach’s α = .72, M = .84, SD = 0.84). Destructive conflict contained four items (Cronbach’s α = .76, M = 1.87, SD = 1.00). Allowances contained five items (Cronbach’s α = .78, M = 2.32, SD = 1.21).
Identity Gaps
Personal-enacted and personal-relational identity gaps (Jung & Hecht, 2004, 2008) are both 12-item scales on a one-seven response format with higher scores reflecting larger identity gaps. Personal-relational identity gaps (e.g., “I am different from the way my communication partners see me”) had a Cronbach α of .85 (M = 2.43, SD = 0.89). For personal-enacted identity gaps (e.g., “I do not reveal important aspects of myself in communication with my communication partners”), Cronbach α was .90 (M = 1.86, SD = 0.79).
Relationship Satisfaction
Relational satisfaction is assessed with the global measure of relational satisfaction (Lawrance & Byers, 1998), which uses a seven-point bipolar scale (e.g., good-bad, valuable-worthless), to assess overall satisfaction with the relationship(s) (Cronbach α = .93, M = 6.30, SD = 1.10). This measure was adapted to assess global satisfaction across multiple relationships, i.e., partner to partner(s).
Resiliency
Resiliency was measured with the six-item resilience scale (Smith et al., 2008) that focuses on resilience as bouncing back, (e.g. “It does not take me long to recover from a stressful event”). Responses fall along a seven-point scale, one = strongly disagree to seven = strongly agree; three items are reverse coded (Cronbach’s α = .92, M = 4.83, SD = 1.34).
Results
To test the four hypotheses, I conduct a series of moderation analyses with each maintenance behavior considered as a separate moderator were conducted with the PROCESS Macro in IBM SPSS (Hayes, 2012). For coherence, I first summarize the results of each hypothesis, then present the findings of each specific analysis. Tables and figures are presented for significant findings.
Hypothesis 1
The first hypothesis predicts that prosocial relational maintenance behaviors will moderate the effects of (a) personal-enacted and (b) personal-relational identity gaps on relationship satisfaction. Findings suggest that H1a is supported for the specific maintenance behaviors of social networks, advice, positivity, openness, and shared tasks, but not supported for conflict management or assurances. Thus, H1a is partially supported. H1b found that none of the prosocial relational maintenance behaviors were supported for a moderating effect on personal-relational maintenance behaviors. Thus, it seems that some prosocial relational maintenance behaviors moderate the effect of personal-enacted identity gaps on relationship satisfaction, but not personal-relational gaps. I describe the findings from each test in this analysis next.
First, personal-enacted identity gaps, assurances, and their interaction do significantly predict relationship satisfaction, F(3, 153) = 21.38, R2 = .24, p < .001, but the interaction term was not statistically significant in contributing to the model (p = .489).
Next, personal-enacted identity gaps, openness, and their interaction do significantly predict relationship satisfaction F(3, 153) = 23.47, R2 = .32, p < .001, and the interaction term was significant in contributing to the model, β = -.20 (SE = 0.09), p = .031), presented in Table 1. At the mean, the raw regression coefficient for openness is -.65 (SE = 0.09), p < .001. At minus-one standard deviation, the regression coefficient for openness is -.50 (SE = 0.09), p < .001; at plus-one standard deviation, the raw regression coefficient is -.81 (SE = 0.13), p < .001. Based on this set of regression analyses, a graphic representation of the interaction effect is presented in Figure 1. It appears when personal-enacted identity gaps are high, openness exacerbates the negative association with relational satisfaction, resulting in increased relational dissatisfaction.
Table 1
Standard Regression Results for the Interaction Based on the Centered Variables for H1a Openness, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 5.522 | 1.204 |
| Openness | 0.330 | 0.194 |
| Personal-enacted identity gaps | 0.577 | 0.536 |
| Openness x P-E identity gaps* | -0.200 | 0.091 |
*Significant at p < .05.
Figure 1
Interaction effects for Openness moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Personal-enacted identity gaps, conflict management, and their interaction do significantly predict relational satisfaction, F(3, 153) = 23.04, R2 = .31, p < .001. However, the interaction term was not significant in contributing to the model (p = .136).
Personal-enacted identity gaps, shared tasks, and their interaction do significantly predict relational satisfaction F(3, 153) = 25.07, R2 = .33, p < .001, and the interaction term was statistically significant in contributing to the model, β = -.21 (SE = 0.07), p = .004, presented in Table 2. At the mean, the raw regression coefficient for shared tasks is -.59 (SE = 0.07), p < .001. At minus-one standard deviation, the regression coefficient for shared tasks is -.38 (SE = 0.09), p < .001; at plus-one standard deviation, it is -.80 (SE = 0.11), p < .001. Based on this set of regression analyses, a graphic representation of the interaction effect is presented in Figure 2. It appears that when personal-enacted identity gaps are high, shared tasks being high strengthens the negative association with relational satisfaction, also resulting in increased relational dissatisfaction.
Table 2
Standard Regression Results for the Interaction Based on the Centered Variables for H1a Shared Tasks, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 5.278 | 0.881 |
| Shared Tasks* | 0.375 | 0.140 |
| Personal-enacted identity gaps | 0.612 | 0.401 |
| Shared Tasks x P-E identity gaps* | -0.206 | 0.070 |
*Significant at p < .05.
Figure 2
Interaction effects for Shared Tasks moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Personal-enacted identity gaps, the prosocial maintenance behavior of positivity, and their interaction significantly predict relational satisfaction F(3, 153) = 27.02, R2 = .35, p < .001, and the interaction term was statistically significant in contributing to the model, β = -.18 (SE = 0.05), p < .001, presented in Table 3. At the mean, the raw regression coefficient for positivity is -.56 (SE = 0.10), p < .001. At minus-one standard deviation, the regression coefficient for positivity is -.37 (SE = 0.08), p < .001; at plus-one standard deviation, the regression coefficient is -.76 (SE = 0.10), p < .001. Based on this set of regression analyses, a graphic representation of the interaction effect is presented in Figure 3. It appears that when personal-enacted identity gaps are high, positivity being high strengthens the negative association with relational satisfaction.
Table 3
Standard Regression Results for the Interaction Based on the Centered Variables for H1a Positivity, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 5.301 | 0.611 |
| Positivity* | 0.385 | 0.107 |
| Personal-enacted identity gaps | 0.411 | 0.287 |
| Positivity x P-E identity gaps* | -0.176 | 0.052 |
*Significant at p < .05.
Figure 3
Interaction effects for Positivity moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Personal-enacted identity gaps, advice, and their interaction also significantly predict relational satisfaction F(3, 153) = 24.20, R2 = .32, p < .001, and the interaction term was statistically significant, β = .12 (SE = 0.05), p = .010, presented in Table 4. At the mean, the raw regression coefficient for advice is -.55 (SE = 0.07), p < .001. At minus-one standard deviation, the regression coefficient is -.72 (SE = 0.09), p < .001; at plus-one standard deviation, the regression coefficient is -.39 (SE = 0.09), p < .001. Based on this series of three regression analyses, a graphic representation of the interaction effect is presented in Figure 4. Unlike the other prosocial maintenance behaviors so far, advice moderates the effects of personal-enacted identity gaps on relational satisfaction in the predicted direction, appearing to weaken the negative association between personal-enacted identity gaps and relational satisfaction.
Table 4
Standard Regression Results for the Interaction Based on the Centered Variables for H1a Advice, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 8.461 | 0.472 |
| Advice* | -0.210 | 0.092 |
| Personal-enacted identity gaps* | -1.147 | 0.238 |
| Advice x P-E identity gaps* | 0.123 | 0.047 |
*Significant at p < .05.
Figure 4
Interaction effects for Advice moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Lastly, personal-enacted identity gaps, social networks, and their interaction significantly predict relational satisfaction, F(3, 153) = 29.28, R2 = .36, p < .001, and the interaction term was significant, β = .11 (SE = 0.03), p = .003, presented in Table 5. At the mean, the raw regression coefficient for social networks is -.42 (SE = 0.07), p < .001. At minus-one standard deviation, the raw regression coefficient for social networks is -.58 (SE = 0.07), p < .001; at plus-one standard deviation, the raw regression coefficient for social networks is -.30 (SE = 0.09), p = .002. Based on this series of regression analyses, a graphic representation of the interaction effect is presented in Figure 5. It appears that social networks weaken the negative relationship between personal-enacted identity gaps and relational satisfaction. Thus, H1a was supported for social networks, advice, positivity, openness, and shared tasks, but not for conflict management or assurances.
Table 5
Standard Regression Results for the Interaction Based on the Centered Variables for H1a Social Networks, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 7.780 | 0.451 |
| Social Networks | -0.100 | 0.084 |
| Personal-enacted identity gaps* | -1.030 | 0.186 |
| Social Networks x P-E identity gaps* | 0.112 | 0.037 |
*Significant at p < .05.
Figure 5
Interaction effects for Social Networks moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
For H1b, personal-relational identity gaps, each prosocial maintenance behavior, and their interactions significantly predict relational satisfaction, assurances: F(3, 153) = 20.43, R2 = .29, p < .001; openness: F(3, 153) = 18.34, R2 = .26, p < .001; conflict management: F(3, 153) = 19.84, R2 = .28, p < .001; shared tasks: F(3, 153) = 18.41, R2 = .27, p < .001; positivity: F(3, 153) = 19.90, R2 = .28, p < .001; advice: F(3, 153) = 20.35, R2 = .29, p < .001; social networks: F(3, 153) = 25.35, R2 = .33, p < .001. However, none of the interaction terms were statistically significant. Thus, H1b was not supported for the moderating role of prosocial relational maintenance behaviors, although they do contribute to predicting relational satisfaction.
Hypothesis 2
The second hypothesis predicts that prosocial relational maintenance behaviors moderate the effects of (a) personal-enacted and (b) personal-relational identity gaps on resilience in polyamorous relationships. H2a was not supported, with none of the interaction terms reaching statistical significance. For H2b, social networks, advice, and openness appear to moderate the relationship between personal-relational identity gaps and resilience. Thus, while prosocial maintenance behaviors do not appear to moderate the relationship between personal-enacted identity gaps and resilience for polyamorous individuals, specific maintenance behaviors (social networks, advice, and openness) may act as moderators for personal-relational identity gaps and resilience for polyamorous individuals. I describe the results of each test for this hypothesis in this section.
Personal-enacted identity gaps, each prosocial maintenance behavior, and their interactions all significantly predict resilience, assurances: F(3, 154) = 4.12, R2 = .07, p = .008; openness: F(3, 154) = 3.55, R2 = .06, p = .016; conflict management: F(3, 154) = 4.45, R2 = .08, p = .004; shared tasks: F(3, 154) = 5.52, R2 = .10, p = .001; positivity: F(3, 154) = 8.51, R2 = .14, p < .001; advice: F(3, 154) = 3.76, R2 = .07, p = .012; social networks: F(3, 154) = 4.24, R2 = .08, p = .007. However, none of the interaction terms were statistically significant.
Thus, H2a was not supported for the moderating role of prosocial relational maintenance behaviors.
For H2b, personal-relational identity gaps, assurances, and their interaction do significantly predict resilience, F(3, 154) = 2.95, R2 = .05, p = .035. However, the interaction term was not a significant contributor to the model (p = .318).
Personal-relational identity gaps, openness, and their interaction significantly predict resilience, F(3, 154) = 4.22, R2 = .08, p = .007. The interaction was a significant contributor to the model, β = .35 (SE = 0.16), p = .034, presented in Table 6. At the mean, the raw regression coefficient for openness is -.29 (SE = 0.14), p = .034. At minus-one standard deviation, the raw regression coefficient for openness is -.57 (SE = 0.18), p = .002; at plus-one standard deviation, the raw regression coefficient for openness is -.02 (SE = 0.19), p = .916. Based on these three regression analyses, a graphic representation of the interaction effect is depicted in Figure 6. Having low levels of openness at low levels of personal-relational identity gaps results in higher resilience, and at high levels of personal-relational identity gaps, low levels of openness result in less resilience. At high levels of openness, it appears that personal-relational identity gaps do not dramatically change resilience, but the result was not statistically significant.
Table 6
Standard Regression Results for the Interaction Based on the Centered Variables for H2b Openness, n = 158
| Variable | B | SEb |
|---|---|---|
| Constant | 10.747 | 2.911 |
| Openness | -0.828 | 0.458 |
| Personal-relational identity gaps* | -2.469 | 1.019 |
| Openness x P-R identity gaps* | 0.352 | 0.165 |
*Significant at p < .05.
Figure 6
Interaction effects for Openness moderating Personal-Relational Identity Gaps on Resilience.
Personal-relational identity gaps, conflict management, and their interaction do significantly predict resilience, F(3, 154) = 3.71, R2 = .07, p = .013. However, the interaction term did not significantly contribute to the model (p = .765). Personal-relational identity gaps, shared tasks, and their interaction do significantly predict resilience, F(3, 154) = 6.46, R2 = .11, p < .001. The interaction term significantly contributed to the model, β = .29 (SE = .13), p = .033, presented in Table 7. However, none of the conditional effects on the centered values of the moderator at the mean (p = .084), minus-one standard deviation (p = .118), and plus-one standard deviation (p = .288) were statistically significant.
Table 7
Standard Regression Results for the Interaction Based on the Centered Variables for H2b Shared Tasks, n = 158
| Variable | B | SEb |
|---|---|---|
| Constant | 8.034 | 2.027 |
| Shared Tasks | -0.415 | 0.334 |
| Personal-relational identity gaps* | -1.979 | 0.795 |
| Shared Tasks x P-R identity gaps* | 0.286 | 0.133 |
*Significant at p < .05.
Personal-relational identity gaps, positivity, and their interaction do significantly predict resilience, F(3, 154) = 8.28, R2 = .14, p < .001. However, the interaction term did not significantly contribute to the model (p = .067).
Personal-relational identity gaps, advice, and their interaction do significantly predict resilience, F(3, 154) = 3.74, R2 = .07, p = .009, and the interaction significantly contributes to the model, β = .19 (SE = 0.09), p = .042, presented in Table 8. At the mean, the raw regression coefficient for advice is -.38 (SE = 0.13), p = .004. At minus-one standard deviation, the raw regression coefficient for advice is .64 (SE = 0.19), p = .001; at plus-one standard deviation, it is -.11 (SE = 0.17), p = .499. Based on these regression analyses, graphic representation of the interaction effect is depicted in Figure 7. Less advice at low levels of personal-relational identity gaps results in higher resilience, and at high levels of personal-relational identity gaps, less advice results in less resilience. With increases in advice, it appears that personal-relational identity gaps do not dramatically change resilience, but the result was not statistically significant.
Table 8
Standard Regression Results for the Interaction Based on the Centered Variables for H2b, Advice, n = 158
| Variable | B | SEb |
|---|---|---|
| Constant | 7.809 | 1.107 |
| Advice* | -0.435 | 0.217 |
| Personal-relational identity gaps* | -1.297 | 0.480 |
| Advice x P-R identity gaps* | 0.191 | 0.093 |
*Significant at p < .05.
Figure 7
Interaction effects for Advice moderating Personal-Relational Identity Gaps on Resilience.
Lastly, personal-relational identity gaps, social networks, and their interaction do significantly predict resilience, F(3, 154) = 4.19, R2 = .08, p = .007. The interaction term significantly contributed to the model, β = .20 (SE = 0.09), p = .034, presented in Table 9. At the mean, the raw regression coefficient for social networks is -.35 (SE = 0.13), p = .006. At minus-one standard deviation, the raw regression coefficient for social networks is -.59 (SE = 0.17), p = .001; at plus-one standard deviation, the raw regression coefficient for social networks is -.12 (SE = 0.16), p = .474. Based on these three regression analyses, a graphic representation of the interaction effect is presented in Figure 8. Lower levels of social networks at low levels of personal-relational identity gaps results in higher resilience, and at high levels of personal-relational identity gaps, less social networks result in less resilience. Thus, H2b was supported for social networks, advice, and openness.
Table 9
Standard Regression Results for the Interaction Based on the Centered Variables for H2b, Social Networks, n = 158
| Variable | B | SEb |
|---|---|---|
| Constant | 7.976 | 1.343 |
| Social Networks | -0.433 | 0.242 |
| Personal-relational identity gaps* | -1.382 | 0.502 |
| Social Networks x P-R identity gaps* | 0.196 | 0.091 |
*Significant at p < .05.
Figure 8
Interaction effects for Social Networks moderating Personal-Relational Identity Gaps on Resilience.
Hypothesis 3
The third hypothesis predicts that antisocial relational maintenance behaviors also moderate the effects of (a) personal-enacted and (b) personal-relational identity gaps on relational satisfaction in polyamorous relationships. H3a was partially supported, with the antisocial maintenance behaviors of allowing control, destructive conflict, and jealousy induction moderating the relationship between personal-enacted identity gaps and relational satisfaction. H3b was also partially supported, but only for the maintenance behavior of spying, with the other antisocial maintenance behaviors (allowing control, destructive conflict, avoidance, and jealousy induction) not acting as moderators for personal-relational identity gaps on relational satisfaction. I describe the results of each test for this hypothesis in this section.
Specifically addressing H3a, personal-enacted identity gaps, jealousy induction, and their interaction do predict relational satisfaction, F(3, 153) = 29.81, R2 = .37, p < .001; the interaction significantly contributes to the model, β = -.35 (SE = 0.10), p < .001, presented in Table 10. At the mean, the raw regression coefficient for jealousy induction was -.50 (SE = 0.07), p < .001. One standard-deviation below the mean was replaced with the minimum value because it would have been outside of the range of possible values for the data-set, at which the raw regression coefficient was -.43 (SE = 0.07), p < .001. At plus-one standard deviation, the raw regression coefficient was -.73 (SE = 0.09), p < .001. Based on this series of three regression analyses, a graphic representation of the interaction is presented in Figure 9. It appears that increases in jealousy induction strengthen the negative relationship between personal-enacted identity gaps and relational satisfaction.
Table 10
Standard Regression Results for the Interaction Based on the Centered Variables for H3a, Jealousy Induction, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 6.690 | 0.283 |
| Jealousy Induction* | 0.567 | 0.213 |
| Personal-enacted identity gaps | -0.074 | 0.138 |
| Jealousy Induction x P-E identity gaps* | -0.354 | 0.095 |
*Significant at p < .05.
Figure 9
Interaction effects for Jealousy Induction moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Personal-enacted identity gaps, avoidance, and their interaction do significantly predict relational satisfaction, F(3, 153) = 21.76, R2 = .30, p < .001. However, the interaction did not significantly contribute to the model (p = .214). Personal-enacted identity gaps, spying, and their interaction also predict relational satisfaction, F(3, 153) = 24.43, R2 = .32, p < .001, but the interaction was not a significant contributor to the model (p = .074).
Personal-enacted identity gaps, destructive conflict, and their interaction do significantly predict relational satisfaction, F(3, 153) = 28.84, R2 = .36, p < .001. In addition, the interaction significantly contributes to the model, β = -.13 (SE = 0.04), p = .004, presented in Table 11. At the mean, the raw regression coefficient for destructive conflict was -.45 (SE = 0.07), p < .001. One standard-deviation below the mean was replaced with the minimum value because it would have been outside of the range of possible values for the data-set, at which the raw regression coefficient was -.34 (SE = 0.09), p < .001. At plus-one standard deviation, the raw regression coefficient was -.58 (SE = 0.08), p < .001. Based on this series of regression analyses, a graphic representation of the interaction is presented in Figure 10. It appears that increases in destructive conflict strengthen the negative relationship between personal-enacted identity gaps and relational satisfaction.
Table 11
Standard Regression Results for the Interaction Based on the Centered Variables for H3a, Destructive Conflict, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 7.054 | 0.237 |
| Destructive Conflict | 0.107 | 0.107 |
| Personal-enacted identity gaps | -0.214 | 0.118 |
| Destructive Conflict x P-E identity gaps* | -0.127 | 0.044 |
*Significant at p < .05.
Figure 10
Interaction effects for Destructive Conflict moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
Lastly, personal-enacted identity gaps, the negative maintenance behavior of allowing control, and their interaction do significantly predict relational satisfaction, F(3, 153) = 28.18, R2 = .38, p < .001. In addition, the interaction significantly contributes to the model, β = -.11 (SE = 0.04), p = .007, presented in Table 12. At the mean, the raw regression coefficient for allowing control is -.39 (SE = 0.08), p < .001. At minus-one standard deviation from the mean, the raw regression coefficient is -.26 (SE = 0.10), p = .014; at plus-one standard deviation, it is -.53 (SE = 0.08), p < .001. Based on this series of regression analyses, a graphic representation of the interaction is presented in Figure 11. It appears that increasingly allowing control strengthens the negative relationship between personal-enacted identity gaps and relational satisfaction. Thus, H3a is supported for allowing control, destructive conflict, and jealousy induction, but not for avoidance or spying, although they do help to explain variance in relational satisfaction.
Table 12
Standard Regression Results for the Interaction Based on the Centered Variables for H3a, Allowing Control, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 6.940 | 0.266 |
| Allowing Control | 0.110 | 0.098 |
| Personal-enacted identity gaps | -0.132 | 0.140 |
| Allowing Control x P-E identity gaps* | -0.113 | 0.041 |
*Significant at p < .05.
Figure 11
Interaction effects for Allowing Control moderating Personal-Enacted Identity Gaps on Relational Satisfaction.
To address H3b, personal-relational identity gaps, jealousy induction, and their interaction do significantly predict relational satisfaction, F(3, 153) = 21.04, R2 = .29, p < .001. However, the interaction does not significantly contribute to the model (p = .389). Personal-relational identity gaps, avoidance, and their interaction do significantly predict relational satisfaction, F(3, 153) = 19.20, R2 = .27, p < .001. However, the interaction does not significantly contribute to the model (p = .158).
Personal-relational identity gaps, spying, and their interaction also significantly predict relational satisfaction, F(3, 153) = 24.82, R2 = .33, p < .001. In addition, the interaction significantly contributes to the model, β = -.13 (SE = 0.06), p = .037, presented in Table 13. At the mean, the raw regression coefficient for spying was -.36 (SE = 0.07), p < .001. One standard-deviation below the mean was replaced with the minimum value because it would have been outside of the range of possible values for the data-set, at which the raw regression coefficient was -.28 (SE = 0.08), p < .001. At plus-one standard deviation, the raw regression coefficient was -.47 (SE = 0.08), p < .001. Based on this series of regression analyses, a graphic representation of the interaction is presented in Figure 12. It appears that increases in spying strengthens the negative relationship between personal-relational identity gaps and relational satisfaction.
Table 13
Standard Regression Results for the Interaction Based on the Centered Variables for H3b, Spying, n = 157
| Variable | B | SEb |
|---|---|---|
| Constant | 7.116 | 0.322 |
| Spying | 0.135 | 0.180 |
| Personal-Relational identity gaps | -0.155 | 0.128 |
| Spying x P-R identity gaps* | -0.129 | 0.061 |
*Significant at p < .05.
Figure 12
Interaction effects for Spying moderating Personal-Relational Identity Gaps on Relational Satisfaction.
Personal-relational identity gaps, destructive conflict, and their interaction do significantly predict relational satisfaction, F(3, 153) = 21.50, R2 = .30, p < .001. However, the interaction did not significantly contribute to the model (p = .866). Lastly, personal-relational identity gaps, allowing control, and their interaction do significantly predict relational satisfaction, F(3, 153) = 24.28, R2 = .32, p < .001. Again, the interaction did not significantly contribute to the model (p = .098). Thus, H3b was only supported for spying, but allowing control, destructive conflict, avoidance, and jealousy induction do not operate as moderators.
Hypothesis 4
The last hypothesis predicts that antisocial relational maintenance behaviors moderate the effects of (a) personal-enacted and (b) personal-relational identity gaps on resilience for individuals in polyamorous relationships. Neither H4a or H4b were supported. While negative maintenance behaviors do contribute to explaining some variance in resilience, they do not appear to moderate the effects of either personal-enacted or personal-relational identity gaps on resilience. I describe the results of each test for this hypothesis in this section.
Specifically addressing H4a, each negative maintenance behavior, personal-enacted identity gaps, and their interactions do significantly predict resilience, jealousy induction: F(3, 154) = 3.89, R2 = .07, p = .01; avoidance: F(3, 154) = 4.83, R2 = .09, p = .003; spying: F(3, 154) = 3.38, R2 = .06, p = .011; destructive conflict: F(3, 154) = 6.36, R2 = .11, p < .001; allowing control: F(3, 154) = 5.31, R2 = .09, p = .002, however, their interactions were not significant. Thus, H4a was not supported, negative maintenance behaviors do not appear to moderate the relationship between personal-enacted identity gaps and resilience, although they do contribute to predicting resilience.
Addressing H4b, personal-relational identity gaps, the negative maintenance behavior of jealousy induction, and their interaction do not significantly predict resilience, F(3, 154) = 2.52, p = .06. Personal-relational identity gaps, and the remainder of the negative maintenance behaviors do significantly predict resilience, avoidance: F(3, 154) = 3.32, R2 = .06, p = .021; spying: F(3, 154) = 3.00, R2 = .05, p = .032; destructive conflict: F(3, 154) = 5.73, R2 = .10, p = .001; allowing control: F(3, 154) = 3.61, R2 = .07, p = .015, but their interactions were not statistically significant contributors to the model. Thus, H4b was not supported; negative maintenance behaviors do not appear to moderate the relationship between personal-relational identity gaps and resilience. However, with the exception of jealousy avoidance, they do contribute to explaining variance in resilience.
Discussion
TRRL contends that when individuals engage in consistent, positive maintenance behaviors, they develop relational reserves that safeguard their relationship from future stressors (Afifi et al., 2016). Identity gaps, or feelings of discrepancy between an internal sense of self and aspects of identity situated within communication and relationships are an uncomfortable occurrence that can represent or potentially manifest in stress (Jung & Hecht, 2004, 2008; Jung et al., 2007). The present study investigated if, as stressors, the experience of personal-enacted and personal-relational identity gaps on relational satisfaction and resilience would be moderated by relational maintenance behaviors in polyamorous relationships. This section first summarizes the present study’s key findings. Next, I explain the findings and contributions from the present study in light of TRRL, maintenance, and polyamorous literature. Finally, I conclude with recommendations for future research and a discussion of the present study’s limitations.
First, the present study hypothesized that repeated use of prosocial maintenance behaviors that demonstrate relational investment may also moderate the effect of identity gaps on relational satisfaction and resilience, which was largely supported. Specifically, social networks, advice, positivity, openness, and shared tasks moderate the effects of personal-enacted identity gaps on relational satisfaction. Advice, social networks, and openness moderate the effects of personal-relational identity gaps on resilience. Allowing control, destructive conflict, and jealousy induction moderate the effects of personal-enacted identity gaps on relational satisfaction. Only spying of the negative maintenance behaviors moderate the effects of personal-relational identity gaps on relational satisfaction. For the most part, as predicted, positive relational maintenance behaviors appear to weaken, and antisocial maintenance behaviors strengthen, the negative association between identity gaps and relational satisfaction and resilience. This is consistent with the predictions of TRRL, which contends that repeated prosocial maintenance behaviors indicate relational investment (Afifi et al., 2016). Repeated investment communication builds reserves that may protect the individual and relationship from the experiences of stressful feelings and events, like identity gaps. Thus, the present study supports this proposition of TRRL by finding that maintenance behaviors do in fact moderate the relationship between an experience of cognitive and affective stress, in a relational type that manages additional stressors, and relational outcomes.
However, several findings worth noting deviate from this prediction or were not significant. Specifically, antisocial maintenance behaviors did not moderate the relationship between either identity gap and resilience, although prosocial maintenance behaviors do. Antisocial maintenance behaviors still contributed to predicting resilience, as suggested by TRRL (Afifi et al., 2016), but it appears that they do not interact with identity gaps in the same way that prosocial maintenance behaviors do in explaining an individual’s resilience. This suggests positive behaviors may be more important to buffering against stressors and building resilience for individuals in polyamorous relationships than negative behaviors. Future research may investigate if this is unique to identity gaps in polyamorous relationships, or if antisocial maintenance behaviors moderate other stressful experiences’ effect on resilience.
In addition, openness, shared tasks, and positivity did moderate the effects of personal-enacted identity gaps on relational satisfaction, but instead of weakening the association as predicted, increases in openness, shared tasks, and positivity at higher levels of personal-enacted identity gaps further decreases relational satisfaction, thus appearing to strengthen the negative association. Although counter to the prediction of TRRL, which the rest of the findings largely support, this may be attributed to the nature of identity gaps. Personal-enacted identity gaps occur within communication (Hecht, 1993; Jung & Hecht, 2004, 2008). Openness, positivity, and shared tasks are communicative maintenance behaviors (Dainton, 2000). One potential explanation for this finding is because the polyamorous community maintains standards for ethical and responsible relational communication that emphasize openness, honesty, and task-sharing among multiple-partners (Sheff, 2014b; Wosick-Correa, 2010). It may be that individuals in this sample felt a pressure to engage in insincere but strategic maintenance behaviors that furthered a discrepancy between their sense of self and the self-performed in communication, thus furthering rather than buffering against the resultant relational dissatisfaction. This would be consistent with the findings for personal-relational identity gaps, as personal-enacted identity gaps refer more to a discrepancy among one’s own communication and sense of self, whereas relational refers more to discrepancies between one’s sense of self and ascriptions made by a communication partner (Jung & Hecht, 2008). Future research is needed to fully investigate this claim, but the findings introduce some insight into the nature of identity-related stressors and relational maintenance in polyamorous relationships.
The present study makes several important contributions to the literature. Polyamorous individuals manage similar and unique relational stressors compared to monogamous individuals (Dixon, 2016). As TRRL notes, relationships have the potential to buffer against or exacerbate the deleterious effects of stress on individual and relational health (Afifi, 2018; Afifi et al., 2016). For individuals in polyamorous relationships, these benefits or risks may be multiplied across multiple relational contexts. The present study’s findings suggest that prosocial maintenance behaviors do, for the most part, serve to weaken the negative association between identity gaps and relational satisfaction and resilience. Scholars and practitioners may leverage this insight to make recommendations for individuals engaged in polyamorous relationships. In addition, the present study demonstrates that TRRL is a viable theory to study issues pertaining to polyamorous relational stressors and relational maintenance and communication. Future research and practice may investigate what these behaviors look like or how those behaviors occur differently in polyamorous relationships compared to monogamous relationships. For instance, if identity gaps emerge because an individual in a polyamorous relationship has not told their parents about all of their partners, how might the remainder of the shared social networks be leveraged to minimize the harmful effects of this identity gap on relational satisfaction and resilience?
Limitations
The present study tests TRRL and largely upholds its propositions in polyamorous relationships, and extends the theory by including identity gaps as a stressful phenomenon. Despite this contribution, the present study should be read in light of several limitations. First, the dependent variable of relational satisfaction had a positive skew, as is not uncommon in relationship research (Bazzini, Stack, Martincin, & Davis, 2007). Second, the perspective in the present study is at an individual level. Testing this at the dyadic and group level in multiple-partner relationships may yield more comprehensive insights. In addition, although relational satisfaction was assessed as an overall sense of satisfaction across relationships, meaningful differences may exist between an individual’s multiple relationships. Future research should collect data at multiple levels to account for the ways that each relationship may experience different effects of relational investment. Lastly, while a survey method allowed for sexual, gender, and geographic diversity, volunteer surveys often reach more educated populations, which may influence the results of the present study.
Conclusion
Despite these limitations, the present study offers a substantial contribution to an important and growing theory as well as data on relational practices for an understudied relational type. TRRL’s contention that relational investments via relational maintenance behaviors safeguard individuals and relationships against stressful experiences by developing relational reserves was tested and largely upheld for individuals in polyamorous relationships. In addition, the variable of identity gaps was introduced as a source of or reflection of individual stress. For a relational type that amplifies the potential sources of relational stressors as a consequence of the possibility for more than one relationship, and thereby amplified potential sources of resilience and support, an investigation into the communication behaviors that may minimize the harmful effects of these stressors and strengthen relationships and resilience warrants continued inquiry. This study offers an initial investigation into this consequential phenomenon for individuals in polyamorous relationships.