A dyadic model of investments: Partner effects on commitment

Abstract

A dyadic approach to studying relationship dynamics yields considerably more insights than examining each partner separately. Yet relatively little research has examined dyadic models of commitment, despite commitment being essential to relationship persistence. Accordingly, we tested a dyadic version of the investment model of commitment. In two cross-sectional studies of couples and one experiment, we tested the role of partner investments and perceived partner investments as novel antecedents of commitment. Studies 1 and 2 demonstrated that greater partner investments were related to greater levels of individuals’ commitment, while controlling for individuals’ own satisfaction with, investments in, and alternatives to the relationship. Study 3 revealed that partner-reported investments predicted commitment independent of perceived partner investments. The findings advance the investment model beyond the individual level, emphasizing the need to examine dyadic elements of relationships.

Keywords

Alternatives APIM commitment couple dyadic investment partner satisfaction

Relationship counselors, marriage coaches, and pop psychology books may give individuals in troubled romantic relationships similar advice when it comes to mending relationships—make time for your partner, put effort into improving the relationship, and engage in mutually enjoyable activities! Such ostensibly simple advice may reveal an underlying truth: investing in a relationship not only strengthens one’s own commitment (Rusbult, 1983) but may strengthen the partner’s commitment as well. Our goal was to examine the way in which individuals’ relationship commitment is dyadically determined through joint contributions from both partners. Previous research has examined other relationship constructs from a dyadic perspective including responsiveness (e.g., Lemay & Neal, 2014; Reis & Clark, 2013), attachment (Brassard, Péloquin, Dupuy, Wright, & Shaver, 2012), gratitude (Kubacka, Finkenauer, Rusbult, & Keijsers, 2011), intimacy (Debrot, Cook, Perrez, & Horn, 2012), and marital quality and satisfaction (e.g., Cobb, Davila, & Bradbury, 2001; Lawrence et al., 2008). We tested a dyadic version of the investment model of commitment (Rusbult, 1983), with a particular focus on how the partner’s investments in a relationship affect individuals’ relationship commitment.

Interdependence theory, commitment, and investments

Interdependence theory (Kelley & Thibaut, 1978) centers on the structure of how two interacting individuals influence each other’s outcomes (e.g., pleasure, pain). Individuals are satisfied with a relationship when the outcomes received from a partner exceed expectations and dependent on a relationship when they would be unable to receive the same outcomes through alternative sources. Rusbult (1980, 1983) developed the investment model of commitment, proposing that together with satisfaction and alternatives, investments in a relationship influence levels of commitment. Investments have been defined as any “resources ‘put into’ a relationship [that] increase the costs of withdrawing from it” (Rusbult, 1980, p. 174). Investments are anything that would be devalued or lost should the relationship end, such as time, self-disclosure, and shared possessions. In contrast to satisfaction and alternatives, which focus on the benefits (and costs) associated with staying in the relationship as a means of increasing dependence and commitment, investments focus on the costs associated with leaving the relationship as a means of increasing dependence and commitment. Rusbult (1980, 1983) proposed that this dependence becomes internalized into a global, subjective feeling of commitment to the relationship that includes psychological attachment, long-term orientation to the relationship, and intent to persist in the relationship (Rusbult, Olson, Davis, & Hannon, 2001). Commitment is one of the best predictors of relationship persistence relative to other predictors such as love or trust (Le, Dove, Agnew, Korn, & Mutso, 2010). Dozens of studies conducted over almost 40 years have provided ample support for the model (Le & Agnew, 2003). Greater satisfaction and investments, and lower alternatives, consistently predict greater commitment.

In turn, commitment is a powerful predictor of pro-relationship behaviors that facilitate relationship persistence; it evokes the thoughts, feelings, and motivation required to perform behaviors that maintain relationships. Such behaviors include perceiving one’s own relationship as superior to others’ relationships (Van Lange & Rusbult, 1995), the tendency to accommodate or forgive negative partner behavior (Rusbult, Verette, Whitney, Slovik, & Lipkus, 1991), devaluing other potential partners (Johnson & Rusbult, 1989; Simpson, Gangestad, & Lerma, 1990), and the willingness to sacrifice for the good of the relationship (Van Lange et al., 1997). Ultimately, such behaviors increase the likelihood that a relationship persists.

Given the importance of commitment in maintaining and enhancing relationships, researchers have explored predictors of commitment beyond satisfaction, alternatives, and investments. Subjective norms, or beliefs about close others’ approval (or disapproval) of behavior, are a unique predictor of commitment when included with the original three investment model predictors (Etcheverry & Agnew, 2004). However, passionate and intimate love, variables from the triangular theory of love (Sternberg, 1986) did not improve the prediction of commitment when added to the investment model of commitment (rather, adding investments to the triangular theory of love increased the amount of variance in commitment the triangular model explained; Panayiotou, 2005). Similarly, investment model predictors mediated, along with commitment, the effects of attachment anxiety and avoidance on pro-relationship behaviors such as accommodation and breakups (Etcheverry, Le, Wu, & Wei, 2013). Investment model predictors also mediated the effect of perceived partner responsiveness on commitment, with perceived partner responsiveness predicting satisfaction, investments, and alternatives, each of which, in turn, predicted commitment (Segal & Fraley, 2016). These studies advanced the research on antecedents and outcomes of commitment, but the theme is that satisfaction, investments, and alternatives consistently explain the lion’s share of the variance in commitment. A logical next step is to extend the analysis to the behavior of the partner.

Dyadic investment model of commitment

Partner effects, defined as the influence of one partner (e.g., personality, behavior) on the other’s outcomes (e.g., personal well-being, relationship commitment; Kenny & Cook, 1999; Kenny, Kashy, & Cook, 2006), are an important component of any thorough understanding of relationship processes. Researchers have examined partner effects by using partner-reported measures or individuals’ perception of the partner to predict individuals’ outcomes. Partner behavior or characteristics may exert a direct influence on individuals’ outcomes. For example, individuals spent less time exploring a new activity and enjoyed it less when their partners reported greater attachment anxiety (Coy, Green, & Davis, 2012), and individuals reported lower relationship satisfaction when their partners reported greater attachment avoidance (Fitzpatrick & Lafontaine, 2017). In addition, individuals’ perception of the partner’s behavior or characteristics may influence individuals’ outcomes (e.g., responsiveness, Lemay & Neal, 2014; Reis & Gable, 2015; self-disclosure, Sprecher & Hendrick, 2004). For example, perceived partner responsiveness is related to higher levels of intimacy and relationship satisfaction (Reis & Gable, 2015). Such findings support the need to examine partner effects within the investment model of commitment.

We sought to examine how partners may influence each other’s commitment. An examination of interdependence theory tenets suggests the potential influence of partner investments on individuals’ commitment above partner alternatives or partner satisfaction. Individuals’ satisfaction and alternatives are subjective and independently determined through experience in previous relationships or comparisons with relationships of friends or acquaintances (Kelley & Thibaut, 1978; Rusbult, 1980). Research suggests that relationship partners differ in their social comparisons (Titus, 1980), with negative interpretations of comparisons predicting less satisfaction and commitment and positive interpretations predicting greater satisfaction and commitment (Morry, Chee, Penniston, & Sucharyna, 2018). These findings underscore that relationship partners make relationship-focused social comparisons independently and that it is their subjective interpretation of these comparisons that influences relationship outcomes. In the only two previous dyadic examinations of the investment model of commitment, Bui, Peplau, and Hill (1996) identified effects for partner alternatives on commitment, but Macher (2013) found no partner effects in one study, and effects for partner satisfaction only in another study. These mixed findings, revealed in largely exploratory, statistically driven approaches suggest that additional research is needed.

We suggest that a stronger, theory-driven approach focused on partner investments may offer additional insights regarding the dyadic nature of the investment model of commitment. Investments generate dependence (and corresponding increases in commitment) by increasing the costs associated with leaving the relationship (Rusbult, 1980, 1983). Partner investments are likely to operate in a similar manner: when a partner invests in the relationship, the result is likely to be an increase in dependence and commitment for both partners. The time, effort, disclosures, and other investments made by one partner are in jeopardy of being lost by both parties. For example, as shared positive memories, sense of identity, and intertwined intellectual lives develop over time, they may be jointly “banked” as part of the shared relationship narrative, increasing commitment. However, the memories in this narrative bank may be tarnished following relationship dissolution. Research on breakups supports this idea, finding changes in self-concept clarity and content (e.g., preferred activities) following a breakup (Slotter, Emery, & Luchies, 2014; Slotter, Gardner, & Finkel, 2010). More tangibly, a partner’s gift may initially generate dependence and commitment for both individuals, but the emotional and (possibly) monetary value associated with the gift may be lost upon dissolution regardless of who retains possession.

Previous research has examined the effect of investment-related partner behaviors on individuals’ commitment, in some cases utilizing models in which partner characteristics and behaviors predict individuals’ commitment indirectly. For example, in the mutual cyclical growth model, the influence of a partner’s pro-relationship behaviors (operationalized as accommodation, sacrifice, and forgiveness) on commitment was mediated by trust (Wieselquist, 2009; Wieselquist, Rusbult, Foster, & Agnew, 1999). Likewise, research on perceived partner responsiveness indicates that satisfaction may primarily mediate the relationship between responsiveness and commitment (Segal & Farley, 2016). In contrast, our model predicts that partner investments will directly influence commitment. Indeed, in other cases, research has identified direct links of partner behaviors with individuals’ commitment. In one study, greater levels of both partner-reported self-disclosure and perceived partner self-disclosure were related to greater commitment, and perceived partner self-disclosure was related to relationship persistence 6 months later (Sprecher & Hendrick, 2004). In addition, research on self-expansion and inclusion of the other in the self (IOS) offers additional support for effects of partner investments (i.e., shared intellectual life, sense of identity), with greater partner IOS predicting greater commitment (Weinstein, Rodriquez, Knee, & Kumashiro, 2016). Each of these (i.e., sacrifice, forgiveness, self-disclosure, sense of identity) has been identified as a form of intangible investments, a particularly strong predictor of commitment (Goodfriend & Agnew, 2008).

Despite the evidence that these partner behaviors affect commitment at a granular level, research is needed to better understand their cumulative effect on commitment within the framework of the investment model of commitment. We know of only one previous publication (Joel, Gordon, Impett, MacDonald, & Keltner, 2013) that explicitly explored partner investments. In one study manipulating recalled partner (vs. own) investments and two daily dairy studies of perceived partner investments, this research found that perceived partner investments predicted commitment via feelings of trust and gratitude. In the present research, we distinguished between partner investments and perceived partner investments and sought to examine their effects on commitment in the context of the full investment model.

Finally, it is worth noting that Goodfriend and Agnew (2008) speculated that, “there might be a difference between those resources individuals are consciously aware of investing and those they are not (e.g., sacrifices vs. an established routine which slowly becomes automatic over time)” (p. 1651). This echoes the idea of relationship processes becoming habitual (Rusbult & Arriaga, 2000). Putting resources into relationships may become less effortful over time, and individuals may come to no longer consider these behaviors as investments. However, such lack of awareness would not necessarily mean that these investments would stop nor would it mean that the investments would no longer influence the commitment of both partners. Although the investor may no longer consciously consider such habitual behavior to be an investment, the partner may still perceive it as such and respond with increased commitment. This notion is supported by research findings that perceived partner disclosure (Sprecher & Hendrick, 2004) and responsiveness (Lemay & Neal, 2014; Reis & Gable, 2015) predict commitment independently from partner-reported behavior. Therefore, as prior research is mixed, it is necessary that we examine both partner-reported investments and perceived partner investments.

In summary, theory (Kelley & Thibaut, 1978; Rusbult, 1980) and prior empirical findings (Joel et al., 2013; Sprecher & Hendrick, 2004) suggest that investments may be particularly likely to exhibit partner effects within a dyadic framework. We propose and test a dyadic investment model that extends previous work (e.g., Reis & Gable, 2015; Wieselquist, 2009) by evaluating the direct effect of partner-reported investments on commitment. We hypothesized that higher levels of partner investments would be associated with higher levels of commitment and tested this via three studies using different methodologies. Study 1 offers a test of partner investments as a direct predictor of commitment. Study 2 provides an experimental test of partner investment effects. Study 3 offers an additional test of the model and examines perception and mediated effects of partner investment. Because both Studies 1 and 3 evaluated dyadic (i.e., couples) data, the actor–partner interdependence model (APIM; Kenny et al., 2006) was employed as the analysis strategy. APIM accounts for the nonindependence of observations when using couples data by factoring out couple-level variability in the data and allows partner effects to be tested.

Study 1

Study 1 examined basic partner effects in the investment model of commitment. Though a central feature of interdependence theory (Kelley & Thibaut, 1978) is that partners affect each other’s outcomes, little empirical work has directly tested the influence partner investments have on commitment while controlling for other factors. We predicted that individuals would report greater commitment to the degree that the partner reported greater partner investments when controlling for individuals’ own satisfaction, investments, and alternatives.

Method

Participants

Seventy-seven married opposite-sex couples (154 participants) from the Southeastern United States were recruited via notices posted on campus, in the community, and in the newspaper.¹ Couples received USD$50 for completing the session, which was part of a larger study on married couples (see Table 1 for additional demographics).

Table 1.

Demographics for the samples in each study.

	Study 1	Study 2	Study 3
Sex
Male	77 (50%)	49 (34%)	91 (45%)
Female	77 (50%)	95 (66%)	110 (54%)
Age	34 years (22–76 years)	19 years (18–48 years)	20 years (18–46 years)
Race/ethnicity
Caucasian	123 (80%)	74 (51%)	98 (49%)
African American	16 (10%)	25 (17%)	67 (33%)
Asian American	3 (2%)	23 (16%)	14 (7%)
Hispanic/Latino/a	6 (4%)	7 (5%)	10 (5%)
Other/not specified	6 (4%)	15 (10%)	13 (6%)
Relationship length	72 months (1–600 months)	17 months (1–60 months)	19 months (2–96 months)
Relationship status
Dating steadily	—	114 (79%)	81 couples (80%)
Dating casually	—	18 (13%)	—
Engaged/married	100%	6 (4%)	8 couples (8%)
Other	—	6 (4%)	12 couples (12%)

Measures

Participants completed all measures individually and were instructed not to discuss their responses. Commitment to the marriage was assessed using an expanded 15-item version of the Investment Model Scale (Rusbult, Martz, & Agnew, 1998) adapted for married couples. The commitment measure assessed long-term orientation, psychological attachment, and intent to persist in the marriage (e.g., “I feel completely attached to my partner and our relationship”; “When I make plans about future events in my life, I carefully consider the impact of my decisions on our marriage”). We measured the antecedents to commitment using the Investment Model Scales (Rusbult et al., 1998), adapted for married couples. Five items measured each antecedent: satisfaction with the marriage (e.g., “Our marriage is close to ideal”), investments in the marriage (e.g., “I have invested a great deal in our marriage, in both material and nonmaterial ways”), and alternatives to the marriage (e.g., “I have acceptable alternatives to our marriage [seeing another person, spending time with friends or on my own]”), on a 9-point scale (0 = do not agree at all; 8 = agree completely). Means, standard deviations (SDs), and correlations for all scales are reported in Table 2.

Table 2.

Means, standard deviations, and intercorrelations among measures in Study 1.

	1.	2.	3.	4.	5.	6.	7.	8.
M	6.89	6.36	4.20	6.79	6.93	6.21	4.13	6.88
SD	1.25	1.10	2.07	1.05	1.31	1.35	2.00	0.99
1. Satisfaction	α = .95
2. Investments	.37**	α = .71
3. Alternatives	−.24	−.26*	α = .89
4. Commitment	.76***	.45***	−.27*	α = .92
5. Partner satisfaction	.71***	.18	−.15	.63***	α = .94
6. Partner investments	.13	.09	−.08	.27*	.30*	α = .83
7. Partner alternatives	−.01	−.05	.20	.04	−.08	−.21	α = .89
8. Partner commitment	.56***	.26*	−.19	.69***	.70***	.45***	−.23	α = .91

***p < .001; **p < .01; *p < .05.

Results

As recommended by Kenny, Kashy, and Cook (2006), to determine whether or not the dyads should be treated as distinguishable, or different based on a particular dimension that differentiates members of the dyad (i.e., sex in the present data), we examined a model with means, variances, and intrapersonal and interpersonal correlations constrained. The χ² test indicated that this model was a good fit to the data, χ²(20) = 13.67, p = .85; therefore, members of the dyads should be treated as indistinguishable (i.e., there were no sex differences in the data) and the path coefficients should be constrained to be equal across partners.

To test the primary hypothesis that higher partner investments would predict greater commitment, we performed an APIM path analysis (Kenny et al., 2006) using Mplus software (Version 6.0) (Muthén & Muthén, 2009), with the mean-corrected maximum likelihood method of parameter estimation to allow for non-normal and missing data, and the full data set as the input (see Table 2 for correlations). Maintaining the constraints described above for the means, variances, and correlations ensures couple-level variability is controlled for in the model. First, the standard investment model (Rusbult, 1980) was modeled for both partners, while constraining the parameter estimates across both models. The model provided a good fit, χ²(18) = 24.29, p = .15; comparative fit index (CFI) = .97; and residual mean square error of approximation (RMSEA) = .07.

Adding the expected relationship between partner investments and commitment significantly improved upon the base investment model, Δχ²(1) = 7.21, p < .01; χ²(17) = 17.08, p = .45; CFI = 1.00; and RMSEA = .01 (see Figure 1). Because the partners were determined not to be distinguishable, the path coefficients are the same for both actors and partners. Critical to the hypothesis, partner investments predicted commitment, βs = .12, p < .05. Alternative models that included direct paths from partner satisfaction and alternatives failed to provide better or theoretically meaningful fit; thus, we retained this model.²

Figure 1.

Path model examining partner effects in the investment model in Study 1. Curved lines represent correlations. Solid lines represent significant unstandardized path coefficients. Paths with the same number in parentheses were constrained to be the same.

Discussion

We confirmed our hypothesis: commitment was predicted not only by the individual’s own satisfaction, investments, and alternatives but also by partner investments. Thus, it seems that partners affect how committed individuals feel via the degree to which the partners invest in the relationship. Put differently, by investing in their relationship (e.g., time, effort, money), one partner may increase the degree to which the other feels committed to the relationship by increasing the extent to which both partners stand to lose should the relationship end. Partner satisfaction and partner alternatives failed to improve model fit, consistent with our theory-driven expectations.

One potential limitation of Study 1 was the exclusion of non-married couples. However, a comprehensive meta-analysis of investment model studies revealed that investments are more highly correlated with commitment for dating relationships than for engaged or married couples (Le & Agnew, 2003). Thus, the focus on married couples actually may have limited the strength of the relationship between partner investments and commitment. Importantly, experimental evidence is needed to make causal inferences about the link between partner investments and commitment.

Study 2

To build on the initial findings from Study 1, Study 2 offered an experimental examination of the relationship between partner investments and commitment. Goodfriend and Agnew (2008) found that investments may be delineated along two dimensions: materiality (i.e., tangible vs. intangible) and timing (i.e., past vs. planned). Intangible investments—consisting of all things immaterial that are put into the relationship (e.g., self-disclosure, effort)—predicted commitment better than tangible investments (e.g., shared monetary investments, pets). Moreover, intangible investments were a particularly powerful predictor of commitment when they were planned (future-oriented as opposed to past investments, e.g., planned date night), and planned sacrifices loaded strongly onto their measure of planned intangible investments. In Study 2, we manipulated future and intangible partner investments in the relationship via sacrifice. Specifically, participants read a short paragraph describing their partners as planning to make many (vs. few) sacrifices and compromises for the individuals’ benefit. We predicted that individuals imagining high future partner investments in their relationship would report greater commitment than individuals imagining low future partner investments. Furthermore, we predicted that this effect would hold even when controlling for individuals’ own levels of satisfaction, investments, and alternatives.

Method

Design

A between-groups design was used to test the hypothesis that partner investments would predict commitment. We randomly assigned participants to read one of two paragraphs about future partner investments (i.e., many or few). Participants also reported their own satisfaction, investments, and alternatives prior to the manipulation, along with their perception of partner investment following the manipulation.

Participants

The number of participants needed was determined using Cohen’s (1992) power primer; with the expected effect size based on the findings of Study 1 and the effect sizes found in related research (e.g., Goodfriend & Agnew, 2008; Sprecher & Hendrick, 2004) and standard power (i.e., .80), this study required 91 participants for a regression model with 5 predictor variables. However, given that one of the variables was a manipulated, categorical variable, and Cohen suggests 64 participants per experimental condition, we estimated a sample size of 150 participants would be adequate to detect effects.

Participants consisted of 177 undergraduate students at a mid-Atlantic university who had been in a dating relationship for at least 1 month and completed the study for partial course credit. Due to concerns related to thoughtless patterned responding and/or true relationship status (i.e., only friends), 33 participants were dropped from the analyses. The final sample consisted of 144 participants (see Table 1 for additional demographics).

Measures

Participants completed the same measures as Study 1 with two modifications. They completed the original Rusbult, Martz, and Agnew’s (1998) Investment Model Scales rather than the versions referencing marital relationships used in Study 1. Participants also completed the 5-item investments measures modified to measure their perception of partner investments following the manipulation as a manipulation check (e.g., “My partner would say that many aspects of his/her life have become linked to mine (recreational activities, etc.), and s/he would lose all of this if we were to break up.”). Means, SDs, and correlations for all scales in Study 2 are reported in Table 3.

Table 3.

Means, standard deviations, and intercorrelations among measures in Study 2.

	1.	2.	3.	4.	5.
M	6.05	4.73	3.81	5.66	5.02
SD	1.58	1.77	1.52	2.25	1.88
1. Satisfaction	α = .92
2. Investments	.38***	α = .85
3. Alternatives	−.18*	−.21*	α = .78
4. Commitment	.49***	.42***	−.26**	α = .95
5. Perceived partner investments	.37***	.49***	−.16	.61***	α = .89

***p < .001; **p < .01; *p < .05.

Manipulation

To manipulate partner investments as either high or low, participants read a brief scenario in which they were asked to imagine that in the future their partners would make many (vs. few) sacrifices and compromises for their benefit (see Appendix A). To make the manipulation more powerful, participants entered their partner’s initials, which were automatically fed into the scenario. Participants then wrote about their specific example and visualized how the scenario would affect the relationship.

Procedure

Participants completed the study via computer and were randomly assigned to either the many (n = 76) or few (n = 68) sacrifice condition. Prior to beginning, participants were asked to ensure that they would not be distracted, and three questions embedded at different points in the survey were used to check for patterned responding.

Participants first completed measures of their own satisfaction, investments, and alternatives. Next, they read either the high or low partner investments manipulation and completed a state measure of partner investments used as a manipulation check. Finally, they then completed the commitment scale and demographic measures.

Results

Prior to testing the hypothesis, we confirmed the effect of the manipulation. Those who imagined low levels of sacrifice perceived their partner as investing less than those who imagined high levels of sacrifice, t(142) = −3.40, p = .001.

To test the hypothesis that individuals considering high partner investments (i.e., many sacrifices) would report greater commitment than participants considering low partner investments (i.e., few sacrifices), we conducted independent samples t-tests and, as expected, individuals who imagined their partner sacrificing little (M = 4.96) reported significantly lower commitment relative to individuals who imagined their partner sacrificing a great deal (M = 6.29), t(142) = −3.62, p < .001.

To examine this effect in the context of the investment model, satisfaction, investments, and alternatives were entered into the model with commitment as the outcome. This model significantly predicted commitment, F(3,140) = 21.64, p < .001, R ² = .32, with satisfaction, β = .36, t(140) = 4.67, p < .001, and investments, β = .34, t(140) = 3.50, p = .001, as significant predictors of commitment, and alternatives as a marginal predictor of commitment, β = −.14, t(140) = −1.93, p = .056. Next, the dummy-coded variable representing the partner investments manipulation was entered. This model explained an additional 6% of the variance in commitment, F(4,139) = 20.93, p < .001, R ² = .38. In addition, manipulated investments remained a significant predictor of commitment, with those in the high partner investment condition reporting greater commitment, β = .24, t(139) = 3.63, p = .001, with satisfaction, β = .35, t(139) = 4.72, p < .001, and investments, β = .24, t(139) = 3.36, p = .001, each remaining significant predictors, and alternatives now a significant predictor, β = −.20, t(139) = −2.02, p = .05, confirming our hypothesis.³

Discussion

Building on the findings of Study 1, Study 2 manipulated partner investments successfully and confirmed that greater partner investment, operationalized as high (vs. low) future partner sacrificing behavior, increases commitment. Taken together, Studies 1 and 2 provided converging empirical support for the notion that an investment in the relationship by one partner may increase the commitment level of both partners.

The manipulation of partner investments involved participants imagining that their partner would make many (vs. few) sacrifices and compromises for the individuals’ benefit in the future. The current finding regarding future partner investments complements Joel and colleagues' (2013) experimental finding that recalling (past) partner investments resulted in greater commitment. Previous research found that individuals’ own intangible investments, including sacrifice, were related to their own commitment regardless of whether they occurred in the past or were planned (would occur in the future; Goodfriend & Agnew, 2008). Future research could continue to examine alternative manipulations of partner investments that vary in timing (past vs. planned) and materiality (tangible vs. intangible).

Although our manipulation focused on the partner’s behavior, any such experimental manipulation is unable to disentangle the partner’s behavior from individuals’ perception of the partner’s behavior. This begs the important question: do people need to perceive (be aware of) partner investments for them to have an effect on commitment? Study 3 sought to address this issue.

Study 3

Studies 1 and 2 both offered evidence that partner investments increase commitment. However, previous research suggests that an individual’s perceptions of partner behaviors and characteristics, including elements of investments (e.g., Joel et al., 2013; Sprecher & Hendrick, 2004), may have unique effects over and above the partner’s behavior. Thus, Study 3 provided an additional test of the dyadic model of investments and also examined the importance of the perception of partner investments.

The perception of partner behavior and partner-reported behavior frequently explain unique variance in relationship outcomes (e.g., Lemay & Neal, 2014; Reis & Gable, 2015). However, findings are mixed regarding the degree to which the perception of partner characteristics and behaviors corresponds with the partner’s actual characteristics and behaviors (Gagné & Lydon, 2004; Kenny & Acitelli, 2001; Reis & Gable, 2015). Some researchers have emphasized perceptions (e.g., Joel et al., 2013; Wieselquist, 2009; Wieselquist et al., 1999), whereas others have found independent effects for both perceptions and partner-reported behaviors (e.g., Cohen, Schulz, Weiss, & Waldinger, 2012; Sprecher & Hendrick, 2004).

In addition, investments may be made habitually or without conscious awareness (Rusbult & Arriaga, 2000), limiting the extent to which they are reported or perceived, but this does not mean that these behaviors do not influence commitment. We suggest the same may be true for partner investments: individuals may become accustomed to a partner’s investment and no longer consciously reflect on them, but they may still influence commitment. Therefore, we predicted that both actual partner investments and perceived partner investments would exert independent effects on commitment.

Method

Participants

Ackerman and Kenny’s (2016) power analysis tool was used to estimate the sample size needed. Using a β of .25 for the expected actor effect and .20 for the expected partner effect, a correlation of .20 for the predictors and a correlation of .40 for the errors (based on the findings from Studies 1, 2, and previous research), approximately 86 couples would be needed to detect effects at an α value of .05 and power of .80. Couples were recruited from the university participant pool and at least one partner was given partial course credit for participating. Partners not in the participant pool could elect to be entered into a drawing for a USD$50 gift card or one of five USD$20 gift cards. One hundred thirty-six couples (272 participants) volunteered. Of those, 35 couples were removed due to concerns regarding one or both partners failing to properly complete attention check questions. The final sample consisted of 101 couples (202 participants): 92 opposite-sex couples and 9 same-sex couples (see Table 1 for additional demographics).

Measures and procedure

Participants completed the Investment Model Scales (Rusbult et al., 1998). To assess the perception of partner investments, participants completed the same scales again with instructions that read, “To what extent does each statement describe your partner’s attitudes about your relationship?” (e.g., “My partner feels very involved in our relationship—like s/he has put a great deal into it.”). Thus, both partners reported their own investments in the relationship as well as their perception of their partner’s investments. All participants in the participant pool completed the study in the lab, whereas their partners were encouraged to attend in person but given the option of completing the survey online shortly after their in-lab partner participated. Means, SDs, and correlations for all scales in Study 3 are reported in Table 4.

Table 4.

Means, standard deviations, and intercorrelations among measures in Study 3.

	1.	2.	3.	4.	5.	6.
M	6.51	3.28	5.55	5.61	5.55	6.07
SD	1.48	1.88	1.56	1.59	1.56	1.21
1. Satisfaction	α = .91
2. Alternatives	−.39***	α = .85
3. Investments	.44***	−.23**	α = .73
4. Perceived partner investments	.47***	−.21**	.73***	α = .79
5. Partner-reported investments	.23***	−.24***	.38***	.38***	α = .73
6. Commitment	.53***	−.53***	.37***	.29***	.30***	α = .79

Note. Investments and partner-reported investments are the same scale but crossed for each partner, therefore the mean, standard deviation, and α coefficient are the same, but the correlations differ.

***p < .001; **p < .01; *p < .05.

Results

Although our sample included same-sex couples, we tested the distinguishability using only the opposite-sex couples via the online Dingy application (Kenny, 2015) that tests for distinguishability and interdependence. The omnibus χ² test indicated that a fully constrained model was a good fit to the data, χ²(30) = 38.54, p = .14, and therefore partners should be treated as indistinguishable (i.e., no sex differences); analyses combining same and opposite-sex couples are reported.

Generalized least squares modeling with a compound symmetry variance–covariance structure in the nlme package in R (Pinheiro, Bates, DebRoy, Sarkar, & R Core Team, 2017) was used to analyze an APIM (Kenny et al., 2006). A null model (i.e., a model with no predictors) revealed that approximately 39% of the variance in commitment was shared across the couple, indicating that controlling for the relationship in the analysis was necessary. We first specified the basic commitment model with satisfaction, t = 5.13, p < .001, alternatives, t = −6.22, p < .001, and investments, t = 1.98, p = .05; each significantly predicted commitment. Next, partner investments was entered into the model and was a marginally significant predictor of commitment, t = 1.78, p = .077. In addition, in this model, investments also was a marginal predictor of commitment, t = 1.82, p = .07, and both satisfaction and alternatives remained as significant predictors. This partner investments model improved on the basic investment model, explaining an additional 3% of the variance in commitment, for a total of 42% of the variance in commitment.⁴ Finally, the perception of partner investments was entered, and it significantly, but inversely, predicted commitment, t = −2.08, p = .04. Both partner-reported investments, t = 2.11, p = .04, and investments, t = 2.73, p = .01, were significant predictors of commitment. To determine whether partner-reported investments and perceived partner investments shared variance in the model resulting in a suppression effect, we tested a model without partner investments. Perceived partner investments was no longer significantly related to commitment (though the relation was still negative), t = −1.73, p = .08.

Discussion

Study 3 offered some additional support for the central finding from the previous studies: partner investments predicted commitment above and beyond individuals’ own levels of satisfaction, alternatives, and investments. When only partner-reported investments were added to the study, the relationship was marginal. However, Study 3 sought to test whether the effect of partner investments on commitment was independent of the perception of partner investments. In this model, partner investments was a significant predictor of commitment, whereas the perception of partner investments had an unexpected, negative relationship with commitment, in spite of the fact that the bivariate correlation was positive (r = .29). Prior research on responsiveness found that individuals tend to project their own responsiveness on their partner and these perceptions are strongly related to satisfaction (Lemay & Clark, 2008; Reis & Gable, 2015; Segal & Fraley, 2016). Consistent with those findings, we note that the bivariate relationship between perceived partner investments and investments was strong (r = .73), particularly when compared to the relationship between perceived partner investments and partner-reported investments (r = .38). In addition, the perception of partner investments was also more strongly associated with satisfaction (r = .47). Thus, the negative relationship in the model is likely due to the shared variance between the perception of partner investments, satisfaction, and investments. Given this, we note that the inclusion of perceived partner investments had little effect on the coefficient for partner-reported investments. Thus, we find it reasonable to conclude that partner investments was indeed an independent predictor of commitment (separate from perceived partner investments), consistent with our hypothesis.⁵ Future research is needed to examine the relationship between perceived partner investments and commitment; the inverse relationship found in Study 3 conflicts with previous research (Joel et al., 2013) and may be spurious.

In summary, investing in a relationship is a more dynamic and dyadic process than previously demonstrated but is in line with the foundational assumptions of interdependence theory and the investment model (Kelley & Thibaut, 1978; Rusbult, 1980, 1983). Investing in a relationship has a “two for one” punch, building commitment in both partners. This increased commitment, in turn, may elicit additional pro-relationship processes (e.g., Rusbult et al., 1991; Van Lange et al., 1997). Thus, overall, the study provides support for a dyadic investment model.

General discussion

Kelley and Thibaut (1978) developed interdependence theory to explain the complex interactions between people. The theory has become a mainstay in the field, particularly through the investment model of commitment (Rusbult, 1980). However, research has only begun to examine these models from a dyadic perspective (e.g., Campbell & Simpson, 2013; Segal & Fraley, 2016). Three studies with varying methodologies provided evidence for the theory-driven influence partners have on individuals’ relationship commitment. Specifically, partner investments directly predicted commitment in cross-sectional studies of exclusively married couples (Study 1) and both dating and married couples (Study 3). Moreover, to examine a possible causal relationship, we manipulated partner investments in a sample of predominantly dating couples (Study 2) and found that greater partner investments corresponded to increased commitment. In addition, Study 3 determined that the relationship between partner investments and commitment is independent of individuals’ perceptions of partner investments. In sum, a dyadic investment model of commitment confirms and specifies the interdependent nature of close relationships and advances our understanding of the multifarious causes of commitment.

It is noteworthy that the theory-driven approach resulted in consistent findings across multiple studies. That is, interdependence theory suggested—and we empirically verified—that partner investments are rather interdependent in nature, meaning that investments by both parties are specific assets put into the relationship for mutual benefit, and both parties experience the loss or degradation of these investments if the relationship terminates. In contrast, both theory and empirical work suggest that satisfaction and alternatives are more subjective and independent judgments (Morry et al., 2018; Rusbult, 1980; Kelley & Thibaut, 1978; Titus, 1980). Although our results differ from previous work that took dyadic, but largely exploratory, approaches (Bui, Peplau, & Hill, 1996; Macher, 2013), the present three studies yielded generally consistent findings for partner effects in the investment model with varying samples/demographics and research designs. Thus, the dyadic investment model offers a novel path forward for better understanding the nature of romantic relationships.

To better place the dyadic investment model in the context of previous research and theory on related models, our model adds to previous findings and theory on responsiveness and the mutual cyclical growth model. Previous research found that the perception of partner responsiveness and pro-relationship behavior, though possibly a projection of one’s own traits (e.g., Lemay & Clark, 2008; Reis & Gable, 2015), contributes to satisfaction and, in turn, commitment (Segal & Fraley, 2016; Wieselquist, 2009; Wieselquist et al., 1999). Only one previous publication (Joel et al., 2013) examined the influence of perceived partner investments on commitment, finding that perceiving that a partner has invested in the relationship increases commitment. Our findings extend this past work in two primary ways. First, we took a dyadic approach by simultaneously evaluating the influence on commitment of investments reported by both relationship partners along with the perception of partner investments. Second, whereas Joel et al. (2013) evaluated daily investment behaviors, the current research examined investments at a global level, and simultaneously with satisfaction and alternatives. Thus, the unique contribution of the dyadic investment model is that both partners contribute to the other’s felt dependence (i.e., commitment) through the investments each makes into the relationship.

We note some potential limitations. The couples’ samples consisted largely of younger, Caucasian, and primarily opposite-sex couples (though the Study 3 sample was reasonably diverse on some dimensions, and the three studies collectively were diverse regarding relationship length, race/ethnicity, etc.). In addition, the couples were willing to volunteer for research on romantic relationships and may have qualitatively better relationships than the typical couple. Examining a greater range of relationship outcomes may be important regarding practical applications of the findings.

One fruitful direction for future work would be a longitudinal study of the long-term effects of partner investments in romantic relationships. We also recommend additional experimental work on different forms of investments. Study 2 only tested one possible form of partner investment (sacrifice). Previous research (Goodfriend & Agnew, 2008) suggests all investments may not be equal: tangible investments may have little effect on commitment, particularly when made in the past (rather than planned for the future), whereas intangible investments may have a stronger effect regardless of past or future timing. The same may be true for partner investments. Finally, future work should consider whether the accuracy in perception of partner investments (i.e., perception- and partner-reported investments) would reveal additional insights into how commitment is developed and maintained. Joel et al. (2013) found that daily perceived partner investments contributed to commitment; a better understanding of how daily perceptions develop into global commitment may be particularly interesting to relationship scientists and counselors.

A primary tenet of interdependence theory is that the choices and behavior of each partner influence relationship outcomes for both. Rusbult’s (1980, 1983; Rusbult & Arriaga, 2000) work offered a primary mechanism though which such an effect would occur: investments. Three studies provided direct evidence that partner investments affect relationships by directly influencing commitment. A dyadic approach to studying relationship commitment reflects the truly interdependent nature of relationships.

Footnotes

Author’s note

Paul E. Etcheverry is now at Facebook, USA.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Study 1 was supported by Templeton Foundation Grant 5158 awarded to Dr Caryl Rusbult.

Open research statement

This research was not pre-registered. The data and materials used in the research are available upon request by emailing coya@usf.edu.

Notes

Appendix A

Instructional manipulation of partner investments for Study 2 (note: material in parentheses offers the wording used in the low investment condition).

Think about your current partner, _____. Imagine that in the future, ______ will make many (few) sacrifices and compromises for your benefit. For example, ______ will often (rarely) do what’s best for you, instead of what’s best for ______. Similarly, ______ will often (rarely) make choices based on what you prefer rather than on what ______ prefers. In fact, 9 out of 10 people will sacrifice and compromise for their partners less (more) than _____ will for you. In other words, _____ will sacrifice and compromise for your benefit more (less) than 90% of people in romantic relationships will for their partner.

Please take a few minutes to imagine what it would feel like for _____ to behave this way in your relationship. Think about at least one specific EXAMPLE of ______ (not) sacrificing or compromising for you, and imagine that this is just one example of the type of thing that ______ will do often.

Please take a few minutes to visualize the details of this scenario about your relationship with _____. Visualize how this would affect you and your relationship with _____. Re-read this scenario as many times as needed so that you are truly able to visualize this scenario before continuing with the study.

References

Ackerman

R. A.

Kenny

D. A.

(2016, December). APIMPowerR: An interactive tool for Actor-Partner Interdependence Model power analysis [Computer software]. Retrieved from https://robert-a-ackerman.shinyapps.io/APIMPowerRdis/

Brassard

Péloquin

Dupuy

Wright

Shaver

P. R.

(2012). Romantic attachment insecurity predicts sexual dissatisfaction in couples seeking marital therapy. Journal of Sex and Marital Therapy, 38, 245–262. doi:10.1080/0092623x.2011.606881

Bui

K. T.

Peplau

L. A.

Hill

C. T.

(1996). Testing the Rusbult model of relationship commitment and stability in a 15-year study of heterosexual couples. Personality and Social Psychology Bulletin, 22, 1244–1257. doi:10.1177/01461672962212005

Campbell

Simpson

J. A.

(2013). The blossoming of relationships science. In Simpson

J. A.

Campbell

(Eds.), The Oxford handbook of close relationships (pp. 3–10). New York, NY: Oxford University Press.

Cobb

R. J.

Davila

Bradbury

T. N.

(2001). Attachment security and marital satisfaction: The role of positive perceptions and social support. Personality and Social Psychology Bulletin, 27, 1131–1143. doi:10.1177/0146167201279006

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155–159. doi:10.1037/0033-2909.112.1.155

Cohen

Schulz

M. S.

Weiss

Waldinger

R. J

. (2012). Eye of the beholder: The individual and dyadic contributions of empathic accuracy and perceived empathic effort to relationship satisfaction. Journal of Family Psychology, 26, 236–245. doi:10.1037/a0027488

Coy

A. E.

Green

J. D.

Davis

J. L.

(2012). With or without you: The impact of partner presence and attachment on exploration. Journal of Experimental Social Psychology, 48, 411–415. doi:10.1016/j.jesp.2011.08.008

Debrot

Cook

W. L.

Perrez

Horn

A. B.

(2012). Deeds matter: Daily enacted responsiveness and intimacy in couples’ daily lives. Journal of Family Psychology, 26, 617–627. doi:10.1037/a0028666.

10.

Etcheverry

P. E.

Agnew

C. R.

(2004). Subjective norms and the prediction of romantic relationship state and fate. Personal Relationships, 11, 409–428. doi:10.1111/j.1475-6811.2004.00090.x

11.

Etcheverry

P. E.

Wei

(2013). Attachment and the investment model: Distal and proximal predictors of relationship commitment, maintenance, and persistence. Personal Relationships, 20, 546–567. doi:10.1111/j.1475-6811.2012.01423.x

12.

Fitzpatrick

Latfontaine

(2017). Attachment, trust, and satisfaction in relationships: Investigating actor, partner, and mediating effects. Personal Relationships, 24, 640–662. doi:10.1111/pere.12203

13.

Gagné

F. M.

Lydon

J. E.

(2004). Bias and accuracy in close relationships: A integrative review. Personality and Social Psychology Review, 8, 322–338. doi:10.1207/s15327957pspr0804_1

14.

Goodfriend

Agnew

C. R.

(2008). Sunken costs and desired plans: Examining different types of investments in close relationships. Personality and Social Psychology Bulletin, 34, 1639–1652. doi:10.1177/0146167208323743

15.

Hannon

P. A.

Finkel

E. J.

Kumashiro

Rusbult

C. E.

(2012). The soothing effects of forgiveness on victims’ and perpetrators’ blood pressure. Personal Relationships, 19, 279–289.

16.

Hannon

P. A.

Rusbult

C. E.

Finkel

E. J.

Kumashiro

M. A.

(2010). In the wake of betrayal: Perpetrator amends, victim forgiveness, and the resolution of betrayal incidents. Personal Relationships, 17, 253–278.

17.

Joel

Gordon

A. M.

Impett

E. A.

MacDonald

Keltner

(2013). The things you do for me: Perceptions of a romantic partner’s investments promote gratitude and commitment. Personality and Social Psychology Bulletin, 39, 1333–1345. doi:10.1177/0146167213497801

18.

Johnson

D. J.

Rusbult

C. E.

(1989). Resisting temptation: Devaluation of alternative partners as a means of maintaining commitment in close relationships. Journal of Personality and Social Psychology, 57, 967–980. doi:10.1037/0022-3514.57.6.967

19.

Kelley

H. H.

Thibaut

J. W.

(1978). Interpersonal relations: A theory of interdependence. New York: John Wiley & Sons.

20.

Kenny

D. A

. (2015). An interactive tool for testing distinguishability and nonindependence in dyadic data [Computer software]. Retrieved from https://davidakenny.shinyapps.io/Dingy/

21.

Kenny

D. A.

Acitelli

L. K.

(2001). Accuracy and bias in the perception of the partner in a close relationship. Journal of Personality and Social Psychology, 80, 439–448. doi:10.1037//0022-3514.90.3.439

22.

Kenny

D. A.

Cook

(1999). Partner effects in relationship research: Conceptual issues, analytic difficulties, and illustrations. Personal Relationships, 6, 433–448. doi:10.1111/j.1475-6811.1999.tb00202.x

23.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis. New York: The Guilford Press.

24.

Kubacka

K. E.

Finkenauer

Rusbult

C. E.

Keijsers

(2011). Maintaining close relationships: Gratitude as a motivator and a detector of maintenance behavior. Personality and Social Psychology Bulletin, 37, 1362–1375. doi:10.1177/0146167211412196

25.

Lawrence

Bunde

Barry

R. A.

Sullivan

K. T.

White

G. A.

Adams

E. E.

… Brock

R. L

. (2008). Partner support and marital satisfaction: Support amount, adequacy, provision, and solicitation. Personal Relationships, 15, 445–463. doi:10.1111/j.1475-6811.2008.00209.x

26.

Agnew

C. R.

(2003). Commitment and its theorized determinants: A meta-analysis of the investment model. Personal Relationships, 10, 37–57. doi:10.1111/1475-6811.00035

27.

Dove

N. L.

Agnew

C. R.

Korn

M. S.

Mutso

A. A.

(2010). Predicting nonmarital romantic relationship dissolution: A meta-analytic synthesis. Personal Relationships, 17, 377–390. doi:10.1111/j.1475-6811.2010.01285.x

28.

Lemay

E. P.

Clark

M. S.

(2008). How the head liberates the heart: Projection of communal responsiveness guides relationship promotion. Journal of Personality and Social Psychology, 94, 647–671. doi:10.1037/0022-3514.94.4.647

29.

Lemay

E. P.

Neal

A. M.

(2014). Accurate and biased perceptions of responsive support predict well-being. Motivation and Emotion, 38, 270–286. doi:10.1007/s11031-013-9381-2

30.

Macher

(2013). Social interdependence in close relationships: The actor–partner-interdependence–investment model (API-IM). European Journal of Social Psychology, 43, 84–96. doi:10.1002/ejsp.1926

31.

Morry

M. M.

Chee

K. C.

Penniston

T. L.

Sucharyna

T. A

. (2018). Relationship social comparisons: Comparison interpretations and attributions as predictors of relationship quality. Journal of Social and Personal Relationships. Advanced online publication. doi:10.1177/0265407518755018

32.

Muthén

L. K.

Muthén

B. O.

(2009). Mplus (Version 6) [Computer Software]. Los Angeles, CA: Muthén & Muthén.

33.

Panayiotou

(2005). Love, commitment, and response to conflict among Cypriot dating couples: Two models, one relationship. International Journal of Psychology, 40, 108–117. doi:10.1080/00207590444000230

34.

Pinheiro

Bates

DebRoy

Sarkar

& R Core Team. (2017). nlme: Linear and nonlinear mixed effects models (R package version 3.1-131). Retrieved from https://CRAN.R-project.org/package=nlme

35.

Reis

H. T.

Clark

M. S.

(2013). Responsiveness. In Simpson

J. A.

Campbell

(Eds.), The Oxford handbook of close relationships (pp. 400–423). New York: Oxford University Press.

36.

Reis

H. T.

Gable

S. L.

(2015). Responsiveness. Current Opinion in Psychology, 1, 67–71. doi:10.1016/j.copsyc.2015.01.001

37.

Rusbult

C. E.

(1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16, 172–186. doi:10.1016/0022-1031(80)90007-4

38.

Rusbult

C. E.

(1983). A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. Journal of Personality and Social Psychology, 45, 101–117. doi:10.1037/0022-3514.45.1.101

39.

Rusbult

C. E.

Arriaga

X. B.

(2000). Interdependence in personal relationships. In Ickes

Duck

(Eds.), The social psychology of personal relationships (pp. 79–108). Chichester, UK: Wiley.

40.

Rusbult

C. E.

Martz

J. M.

Agnew

C. R.

(1998). The investment model scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships, 5, 357–391. doi:10.1177/0146167295216002

41.

Rusbult

C. E.

Olson

Davis

J. L.

Hannon

M. A.

(2001). Commitment and relationship maintenance mechanisms. In Harvey

J. H.

Wenzel

(Eds.), Close relationships: Maintenance and enhancement (pp. 87–113). Mahwah, NJ: Erlbaum.

42.

Rusbult

C. E.

Verette

Whitney

G. A.

Slovik

L. F.

Lipkus

(1991). Accommodation processes in close relationships: Theory and preliminary empirical evidence. Journal of Personality and Social Psychology, 60, 53–78. doi:10.1037/0022-3514.60.1.53

43.

Segal

Fraley

R. C.

(2016). Broadening the investment model: An intensive longitudinal study on attachment and perceived partner responsiveness in commitment dynamics. Journal of Social and Personal Relationships, 33, 581–599. doi:10.1177/0265407515584493

44.

Simpson

J. A.

Gangestad

S. W.

Lerma

(1990). Perception of physical attractiveness: Mechanisms involved in the maintenance of romantic relationships. Journal of Personality and Social Psychology, 59, 1192–1201. doi:10.1037/0022-3514.59.6.1192

45.

Slotter

E. B.

Emery

L. F.

Luchies

L. B.

(2014). Me after you: Partner influence and individual effort predict rejection of self-aspects and self-concept clarity after relationship dissolution. Personality and Social Psychology Bulletin, 40, 831–844. doi:10.1177/0146167214528992

46.

Slotter

E. B.

Gardner

W. L.

Finkel

E. J.

(2010). Who am I without you? The influence of romantic breakup on the self-concept. Personality and Social Psychology Bulletin, 36, 147–160. doi:10.1177/014616720935225

47.

Sprecher

Hendrick

S. S.

(2004). Self-disclosure in intimate relationships: Associations with individual and relationship characteristics over time. Journal of Social and Clinical Psychology, 23, 857–877. doi:10.1521/jscp.23.6.857.54803

48.

Sternberg

R. J.

(1986). A triangular theory of love. Psychological Review, 93, 119–135. doi:10.1037/0033-295X.93.2.119

49.

Titus

S. L.

(1980). A function of friendship: Social comparisons as a frame of reference for marriage. Human Relations, 33, 409–431. doi:10.1177/001872678003300605

50.

Van Lange

P. A. M.

Rusbult

C. E.

(1995). My relationship is better than—and not as bad as—yours is: The perception of superiority in close relationships. Personality and Social Psychology Bulletin, 21, 32–44. doi:10.1177/0146167295211005

51.

Van Lange

P. A. M.

Rusbult

C. E.

Drigotas

S. M.

Arriaga

X. B.

Witcher

B. S.

Cox

C. L.

(1997). Willingness to sacrifice in close relationships. Journal of Personality and Social Psychology, 72, 1373–1395. doi:10.1037/0022-3514.72.6.1373

52.

Weinstein

Rodriquez

L. M.

Knee

Kumashiro

(2016). Self-determined self-other overlap: Interacting effects on partners’ perceptions of support and well-being in close relationships. Journal of Research in Personality, 65, 130–139. doi:10.1016/j.jrp.2016.10.011

53.

Wieselquist

(2009). Interpersonal forgiveness, trust, and the investment model of commitment. Journal of Social and Personal Relationships, 26, 531–548. doi:10.1177/0265407509347931

54.

Wieselquist

Rusbult

C. E.

Foster

C. A.

Agnew

C. R.

(1999). Commitment, pro-relationship behavior, and trust in close relationships. Journal of Personality and Social Psychology, 77, 942–966. doi:10.1037/0022-3514.77.5.942