Accurate and biased perceptions of partner’s conflict behaviors shape emotional experience

Abstract

The present study examined both accuracy and bias in perceptions of romantic partners’ destructive behavior during relationship conflicts, as well as implications of accuracy and bias for emotional experience. Perceptions of partners’ destructive behavior during relationship conflicts were associated with indicators’ of partners’ behavior (i.e., target self-reports and objective coder ratings), suggesting that people were somewhat accurate in perceiving their partner’s destructive behavior. Independently of this accuracy effect, the more participants valued their relationship with their partner, the less destructive they interpreted their partner’s behavior to be during the conflict discussion, suggesting biases in perceptions of destructive behavior. These perceptions of destructive behavior, in turn, predicted emotional experience during the conflict. Results suggest that emotional experience during romantic relationship conflicts is dependent on both accurate and biased perceptions of partners’ behavior.

Keywords

Accuracy/bias conflict destructive behavior emotion romantic relationships

Given that conflict occurs in most romantic relationships (McGonagle, Kessler, & Schilling, 1992), it should be recognized that it is not the presence of conflict that distinguishes healthy relationships from unhealthy ones. Rather, the success of relationships is distinguished by how relationship partners respond to the conflict (Canary, Cupach, & Messman, 1995). While some couples may respond to conflict with constructive solutions (Braiker & Kelley, 1979), others may respond more negatively, such as by being angry (Simpson, Rholes, & Phillips, 1996), hostile (Baron et al., 2007; Gottman, 1993; Simpson et al., 1996), distant (Simpson, Rholes, & Winterheld, 2010), insulting (Peterson & DeHart, 2014), or rejecting (Downey, Freitas, Michaelis, & Khouri, 1998) toward one another. These destructive conflict behaviors are problematic for romantic relationships because they can elicit interpersonal threats that undermine trust and satisfaction, such as perceptions of being devalued or feeling unsafe (Murray, Bellavia, Rose, & Griffin, 2003). For example, people interpret their partner’s anger as an indication of their partner’s low relationship commitment, thus leading them to reciprocate these feelings of anger to their partner (Lemay, Overall, & Clark, 2012). Not surprisingly, enacting these negative behaviors during conflicts is associated with lower relationship satisfaction and more negative perceptions of relationship quality (Baron et al., 2007; Gottman, 1993).

Given how threatening these destructive behaviors may be to relationships, some individuals may be reluctant to acknowledge their partner’s destructive behaviors and instead see their partner’s behavior in a more positive light. This positive outlook may be more likely when people especially value their romantic relationships, given that people who value their relationships with their partners are motivated to perceive those partners as caring and accepting (Clark & Lemay, 2010; Lemay, Clark, & Feeney, 2007; Lemay & Melville, 2014; Lemay & Neal, 2013). Thus, through a process of wishful thinking (Kunda, 1990; Murray, Holmes, & Griffin, 1996b), people who strongly value maintaining a close relationship with their partner (i.e., those who are highly committed to their romantic relationship and strongly care for and have positive regard for their partner), may leave their conflict interactions with positively distorted perceptions of their partner’s behavior. Further, this tendency may be associated with emotional experiences during conflicts, consistent with arguments that cognitive appraisals of situations shape emotional experience (Gross, 2002; Gross & John, 2003; Roseman, Spindel, & Jose, 1990).

In the current research, we present a model of emotional experience as a result of accurate and biased perceptions of partner behavior, which is presented in Figure 1. According to this model, perceptions of a partner’s destructive conflict behaviors are partly rooted in reality (i.e., the partner’s actual behavior), but they are also predicted by the extent to which people value a close relationship with their partner. Further, the perceiver’s perceptions of their partner’s destructive behaviors are thought to impact the emotions that perceivers feel during a conflict with their partner. Each path is described in more detail in the following sections.

Figure 1.

Model of emotional experience as a function of accuracy and bias in perceptions of partner’s behavior.

Path A: People accurately perceive their partner’s destructive behaviors

Several studies have demonstrated that people are accurate in their detection of their romantic partner’s thoughts and feelings (e.g., Fletcher & Kerr, 2010; Kenny & Acitelli, 2001), as well as their partner’s responsiveness and support (Abbey, Andrews, & Halman, 1995; Coriell & Cohen, 1995; Lemay et al., 2007). For example, perceptions of a partner’s responsiveness when discussing a personal problem are somewhat accurate, as they are predicted by objective observers’ ratings of the partner’s behavior (Collins & Feeney, 2000; Lemay & Neal, 2013).

Particularly relevant to the current research, Segrin, Hanzal, and Domschke (2009) found that marital spouses were somewhat accurate in their perceptions of each other’s general behavioral styles during relationship conflicts. This held true for all conflict styles, including tendencies to enact negative behaviors, such as explosive expressions of anger, losing control, and withdrawal. Therefore, it is expected that people are somewhat accurate in their perceptions of their partner’s destructive behaviors during conflict discussions (Path A). The current research extends prior work by examining accuracy and bias with regard to a partner’s enactment of behaviors in a specific conflict interaction, rather than generalized conflict styles, and by examining implications of accuracy and bias for emotional experience.

Path B: Perceiver’s partner valuing biases perceptions of partner’s destructive behaviors

Some relationship perceptions, such as being the target of a romantic partner’s destructive conflict behaviors (e.g., hostility), may be so threatening that people are unwilling to face reality, and so they instead appraise the behavior as being less negative than it is. Indeed, research on empathic accuracy in romantic relationships suggests that people are especially inaccurate at reading their partner’s minds when accuracy would be threatening to the relationship (Simpson, Ickes, & Blackstone, 1995). Research on positive illusions in relationships similarly suggests that people see their partner’s traits as more closely aligned with their desires than is warranted by reality (Murray, Holmes, & Griffin, 1996a). These positive biases may stem, in part, from a motivation to maintain relationships. People who are more motivated to maintain a close relationship with their partner have more to gain from the prospect of a high-quality relationship and more to lose from the prospect of relationship deterioration. Hence, relationship perceptions should be positively biased especially when people are highly invested in maintaining their relationship or when they strongly value their relationship with their partner (Clark & Lemay, 2010; Lemay, 2014; Lemay & Melville, 2014; Lemay & Neal, 2013; Rusbult, Van Lange, Wildschut, Yovetich, & Verette, 2000). For example, people remember their partner as more responsive to them on days when they particularly value their partner (Lemay & Neal, 2013), and these motivated biases of partner responsiveness have been found even after controlling for partner’s actual responsiveness as assessed by objective observers or partners themselves (Lemay & Clark, 2008; Lemay & Neal, 2014). More relevant to the present research, people exhibit positive biases in their perceptions of relationships especially when accurate perceptions may threaten relationships (Lemay, Lin, & Muir, 2015; Rusbult et al., 2000). Given that people who strongly value their relationships want to see their partners as caring and loving (Clark & Lemay, 2010; Lemay et al., 2007; Lemay & Melville, 2014; Lemay & Neal, 2013), we predicted that people who strongly value their relationships will underestimate their partner’s destructive behavior, relative to how it is perceived by objective observers and partners themselves (Path B).

Paths C-E: Perceptions of destructive behavior and emotional experience

Perceptions of a romantic partner’s destructive behaviors can serve as a detriment to the relationship. For example, hostility predicts low relationship satisfaction, both for hostile people as well as their romantic partners, and is also associated with greater relationship conflict (MacKenzie et al., 2014). Romantic partners may feel hurt and devalued when this hostility is directed at them (Leary, Springer, Negel, Ansell, & Evans, 1998; Lemay et al., 2012). Therefore, it was expected that perceptions of a partner’s destructive behaviors would predict perceiver’s negative affect (Path C). By the same logic, it was expected that a partner’s actual destructive behaviors would positively predict perceiver’s negative affect (Path D), but that perceivers’ detection of their partner’s actual destructive behavior explains this effect. In other words, people with partners who enact destructive behaviors are expected to experience more negative affect because they accurately detect these destructive behaviors, suggesting an indirect effect of partner behavior on emotion via accurate detection of that behavior (Path A × C).

Notably, the positively biased perceptions of partners that people form in valued relationships may reduce the experience of negative emotions during conflict situations. Consistent with this possibility, prior research on attributions in relationships suggests that making positive attributions for romantic partners’ behaviors is associated with higher relationship quality and more positive emotions in relationships (Bradbury & Fincham, 1990; Fincham, Paleari, & Regalia, 2002). Similarly, past research has found that positively biased perceptions of behaviors enacted by valued romantic partners predict personal well-being, such as reduced negative affect and increased positive affect (Lemay & Clark, 2015; Lemay & Neal, 2013; Lemay & Neal, 2014). However, this past research did not examine perceptions of destructive conflict behaviors or the role of these perceptions in the context of relationship conflict. Thus, in the present research, we expected that people who value their partners will experience less negative affect during the conflict (Path E) and that this effect would be explained by reduced perceptions of their partner’s destructive behavior, controlling for indicators of their partner’s actual behavior (i.e., observers’ reports and the partner’s self-reports). In other words, people who value their partners may experience less negative emotion during conflicts because they see their partners as engaging in less destructive behavior, an indirect effect of partner valuing on emotional experience via perceptions of the partner’s behavior (Path B × C).

Research overview

The model depicted in Figure 1 was tested in the current study of dyadic conflict interactions. In this model, accuracy is suggested by the effect of a target partner’s destructive behavior, as reported by objective observers and partners themselves, on perceivers’ perceptions of the target partner’s destructive behavior during the conflict discussion (Path A). Independently of the target partner’s destructive behavior, perceivers’ valuing of their relationship with their target partner was expected to predict lower perceptions of the target partner’s destructive behavior during the conflict discussion, suggesting bias (Path B). Further, the resulting perceptions of the target partner’s destructive behavior during the conflict discussion, in turn, were expected to predict perceivers’ emotional experience during the conflict (Path C). In addition, the target partner’s destructive behavior was expected to positively predict perceivers’ negative affect (Path D), and perceivers’ valuing of targets was expected to predict less negative affect (Path E), and these effects were expected to be reduced after controlling for perceivers’ perceptions of targets’ behavior. This pattern would suggest that perceivers with partners who enact destructive behavior experience more negative affect because they accurately detect that behavior, and perceivers who strongly value targets experience less negative affect because they tend to perceive targets’ behavior as less destructive. That is, both effects should be mediated by participants’ perceptions of their partner’s destructive behavior during the conflict, suggesting that both accurate detection and biased distortion of partner behavior contribute to emotional experience during relationship conflict. The AxC indirect effect suggests that the partner’s destructive behavior indirectly predicts more negative affect via participants’ detection of that behavior (i.e., accurate perceptions of destructive behavior are expected to shape emotional experience). The BxC indirect effect indicates that valuing the relationship with one’s partner indirectly predicts reduced negative affect via reduced perceptions of the partner’s destructive behavior (i.e., biased perceptions of the partner’s destructive behavior are also expected to shape emotional experience).

No previous studies have directly examined the emotional implications of positively biased perceptions of partners in the context of relationship conflict. Thus, the current model presents a new conception of how positive illusions may shape affective experience within the context of romantic conflict. These processes are important to understand given the privileged role of romantic partners as elicitors of emotion (Berscheid & Ammazzalorso, 2004) and the universality of romantic relationship conflict (McGonagle et al., 1992).

Method

Participants

Participants included both relationship partners in 116 heterosexual romantic couples, M (age) = 22.11 years; SD = 5.99. Due to technical issues, video data were not available for four of our couples, and survey responses made by one participant were lost. Using the guidelines set forth by Scherbaum and Ferreter (2009) for estimating statistical power, assuming a medium effect size and using an intraclass correlation of .47, the power for our sample size is .88.¹ Participants were recruited from a university campus, as well as through community advertisements. Relationships were predominantly dating relationships (78%) or engaged and marital relationships (17.8%). Participants were compensated US$15 or received partial fulfillment of course requirements in their psychology course in exchange for their participation in the 1-hr lab session. The majority of participants were Caucasian (83.7%), but also included participants who were Asian (5%), African American (5.9%), and Hispanic (1.8%).

Procedure

Participants arrived at the laboratory with their romantic partners. They were separated to complete the pre-interaction measures described in the following (in addition to measures not relevant to this manuscript). Then, participants completed the conflict task. The procedures were adapted from prior research on observed conflict interactions (Gottman, 1979; Simpson et al., 1996). Participants were asked to independently generate three to five issues that were a source of significant and unresolved conflict in their relationship with their partner. Using the lists generated by both couple members, the couple members jointly identified the most significant unresolved problem. The researcher then instructed the couples to discuss this issue, giving them the following instructions:

Remember what you were arguing about and why you were upset with your partner. Remember what you were thinking about and how you felt during the argument. After remembering these things, we would like each of you to tell the other what is it about his or her attitudes, habits, or behaviors that bothers you.

Participants were then given 10 min to discuss the issue with their partner while being video recorded. These videos were later rated by a panel of nine coders to assess each participant’s behavior during the conflict discussion. After the interaction, participants returned to separate rooms and completed an additional questionnaire containing the post-interaction measures described in the following. Finally, they were fully debriefed about the purposes of the study.

Pre-interaction measures

Valuing of partners

Participants completed three measures assessing the extent to which they valued a close relationship with their partner, including relationship commitment (e.g., “I am committed to maintaining my relationship with him/her”; 5 items; Cronbach’s α = .86), care for their partner (e.g., “I care for his/her needs”; 5 items; Cronbach’s α = .77), and regard for their partner (e.g., “He/she has a number of good qualities”; 5 items; Cronbach’s α = .83). These items were adapted from other measures of care, commitment, and self-evaluation (Mills, Clark, Ford, & Johnson, 2004; Rosenberg, 1965; Rusbult, Martz, & Agnew, 1998). Items were completed using 9-point scales (1: extremely disagree; 9: extremely agree). Scores on the three measures were strongly correlated (Cronbach’s α = .85), and they were averaged to create a composite index of chronic valuing of partners for use in the primary analyses. High scores on this composite index represent desires to maintain a close relationship with the partner, consistent with prior research using each of these constructs as an indicator that one wants a close relationship (e.g., Lemay, 2014; Lemay & Neal, 2013; Murray, Holmes, & Griffin, 2000; Rusbult, 1983).

Perceptions of partner’s chronic valuing

For ancillary analyses, we also measured perceptions of partner valuing. Participants completed items analogous to the aforementioned chronic valuing items assessing perceptions of their partner’s commitment (e.g., “He/she is committed to maintaining our relationship”; Cronbach’s α = .83), care (e.g., “He/she cares for my needs”; Cronbach’s α = .84), and regard (e.g., “He/she thinks I have a number of good qualities”; Cronbach’s α = .83). Items were completed using the same response scales. Once again, scores on these measures were strongly correlated (Cronbach’s α = .82) and they were averaged to create a composite index of chronic security in the partner’s valuing.

Post-interaction measures

Perceptions of partner’s and own destructive behavior

After the conflict interaction, participants completed 9 items adapted from prior research on negative direct conflict behavior that assessed perceptions of the partner’s destructive conflict behaviors during the interaction (i.e., “How hostile or angry was he/she toward you?,” “How critical or insulting was he/she toward you?,” “To what extent did he/she blame the problem on you?,” “To what extent did he/she threaten you or give you an ultimatum to get you to change?,” “To what extent did he/she demand that you think, feel, or behave in a certain way?,” “How controlling was he/she?,” “How cold or distant was he/she toward you?,” “How rejecting was he/she toward you?,” and “To what extent did he/she ignore or interrupt you?”; Overall, Fletcher, Simpson, & Sibley, 2009). Responses were made using 9-point scales (1: not at all; 9: extremely) and were averaged to create an index of perceptions of the partner’s destructive conflict behavior (Cronbach’s α = .90). In addition, participants completed an analogous set of 9 items to assess perceptions of their own destructive conflict behavior during the interaction on the same behavioral dimensions (e.g., “How hostile or angry were you toward him/her?”; Cronbach’s α = .86) which utilized the same response scales.

Negative affect

After the interaction, participants indicated the extent to which they experienced eight negative emotions during the interaction (“hostile,” “angry,” “upset,” “sad,” “anxious,” “hurt,” “rejected,” and “stressed”). Two of the items (i.e., hostile and upset) overlap with the PANAS (Watson, Clark, & Tellegen, 1988) and the remaining items were created for the present research to be more directly applicable to emotions felt during a conflict situation. These items were rated using 9-point response scales (1: not at all; 9: extremely). Responses for each emotion were averaged together (Cronbach’s α = .90). Higher values indicate more negative affect experienced during the interaction.

Observers’ ratings of interactions

To assess coders’ perceptions of destructive conflict behavior, a panel of nine coders viewed the recorded interactions and rated each participant’s behavior using items and response scales that were analogous to the nine destructive conflict behavior items described earlier (e.g., “How critical or insulting was the participant toward his/her partner?”). Inter-coder reliability was high for each item (intraclass correlation coefficients ranged from .75 to .90; M = .80). Ratings on the same item by the multiple coders were averaged across the coders. In addition, ratings were averaged across the 9 items assessing destructive behavior to create an objective index of participants’ destructive conflict behavior during the interaction (Cronbach’s α = .92). All nine coders provided ratings for every recorded interaction. To minimize potential carryover effects of the partner’s behavior, each video that the coders rated displayed only one of the dyad members, and each coder rated all of the videos in a unique, randomized order to ensure that both dyad members’ videos were not typically coded consecutively. Observers’ ratings of participants’ destructive conflict behavior were strongly correlated with participants’ self-reports of their own destructive conflict behaviors, r(222) = .49, p < .01, suggesting that these two indicators of the partner’s behavior may capture related but somewhat unique aspects of the partner's behavior. Given that these perspectives may provide unique but related information and that there may be reasons for inaccuracies in each measure, these scores were averaged to create a composite index of participants’ actual destructive behavior for use in all analyses except where otherwise specified. This is consistent with approaches in personality research, which utilize multiple accuracy benchmarks to comprehensively index reality (Funder, 1995).

Results

Analysis strategy

The responses provided by each member of the same romantic dyad are not independent. Therefore, we used multilevel models to test predictions while modeling their interdependence. Analyses were conducted using the Mixed Linear Models command in SPSS. The models treated the two dyad members as nested within the same dyad and specified a compound symmetry error structure to account for dyadic interdependence (Kenny, Kashy, & Cook, 2006). This error structure models the covariance across the two partners in the residuals. Predictors were modeled as fixed due to the limited degrees of freedom in dyadic research (see Kenny et al., 2006). Given that we examined the effects of between-subjects variation in the predictor variables, we did not center any of the predictors in the analyses. In addition to unstandardized coefficients, standardized coefficients (βs) are provided as estimates of effect size. Results are described in terms of “perceivers” and “targets” to facilitate comprehension, but it is important to note that each participant served as both perceiver and target. Descriptive statistics for the model variables and a correlation matrix of these variables are presented in Table 1.

Table 1.

Correlation matrix and descriptive statistics of model variables.

Predictor	1	2	3	4	M	SD
1. Perception of own behavior	—	—	—	—	1.92	1.02
2. Perceptions of partner’s behavior	.59***	—	—	—	1.90	1.18
3. Coders’ perceptions of behavior	.49***	.64***	—	—	1.87	0.76
4. Partner valuing	−.26***	−.31***	−.17*	—	7.76	0.92
5. Negative affect	.32***	.50***	.37***	−.24***	2.55	1.37

Note. SD = standard deviation.

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

Accuracy and bias in perceptions of destructive partner behavior

The first analysis tested predictions that perceivers’ perceptions of the target partner’s destructive conflict behavior would reflect a mix of both accuracy and bias. Perceivers’ perceptions of the target partner’s destructive behavior were entered as the outcome variable, and the two predictors were perceivers’ valuing of partners and the index of the target partners’ destructive behavior (the average of coders’ ratings of the target partner’s behavior and the target partner’s self-reports of their own behavior), plus gender and relationship length were included as covariates. Consistent with predictions, perceivers’ perceptions were predicted by target partners’ actual behavior, b = 1.09, β = .71, t = 16.01, p < .001, suggesting accuracy (Path A), and by perceivers’ valuing of the target partner, b = −.16, β = −.12, t = −2.82, p < .01, suggesting bias (Path B). Therefore, these results indicate that perceivers’ perceptions of the target partner’s destructive behavior are derived from a blend of both accuracy and bias. Neither gender nor relationship length moderated the effect of target partners’ actual behavior, ps > .69, nor the effect of perceivers’ valuing of the target partner, ps > .53, on perceivers’ perceptions.

Analysis of discrepancies

The analyses described earlier suggest something about the process of perceiving destructive conflict behavior, but they do not reveal content. In other words, they do not reveal whether perceivers overestimate or underestimate their partner’s destructive behavior as a function of their value for their partner. Therefore, separate analyses were conducted to examine how perceivers’ value for their relationship was related to their discrepancies in perception (i.e., the discrepancy between their partner’s actual destructive behavior and perceivers’ perceptions of that behavior). Perceivers’ perceptions were centered on the index of the partner’s actual destructive behavior by subtracting partner’s actual destructive behavior. As a result, positive values indicate exaggeration of destructive behavior, and negative values indicate underestimation of destructive behavior. This recentered perception index was then regressed on perceivers’ valuing of partners. The intercept in this model indicates the average judgmental discrepancy when perceivers’ valuing has a score of zero (Lemay, Pruchno, & Feild, 2006). To estimate the average discrepancy at low and high levels of perceivers’ valuing of their partner the intercept was examined after perceivers’ valuing was recentered so zero reflected 1 standard deviation (SD) below (“low” valuing) and above (“high” valuing) the mean. When perceivers’ valuing was low, the average discrepancy was positive and significantly greater than zero, b = .19, t = 2.79, p < .01, suggesting overestimation of destructive behavior. When perceivers’ valuing was high, the average discrepancy was negative and significantly lower than zero, b = −.17, t = −2.46, p < .05, suggesting underestimation of destructive behavior. These results indicate that perceivers overestimated the target partner’s destructive behavior following the interaction when their chronic sentiments toward the target partner were relatively negative, and they underestimated the target partner’s destructive behavior when their chronic sentiments toward the target partner were highly positive.

Implications of accuracy and bias for emotion

The next set of analyses tested a mediation model regarding the emotional implications of accuracy and bias. This model posited that perceivers’ perceptions of the target partner’s destructive conflict behavior most directly predict perceivers’ negative emotions during the interaction and mediate the effects of perceivers’ valuing of the target partner and the target partner’s actual destructive behavior on perceivers’ emotions. In other words, this model proposes that both accurate and biased perceptions of the target partner’s destructive behavior predict perceivers’ emotions. Results are displayed in Figure 2. Paths A and B depict the significant accuracy and bias effects, respectively, which were described earlier.

Figure 2.

Contribution of perceivers’ accuracy (regarding their partner’s destructive behavior) and bias (perceivers’ valuing of partner) to the experience of negative emotions during the relationship conflict discussion. Note. Coefficients are unstandardized. Coefficients in parentheses depict effects without controlling for perceptions of partners’ destructive behavior. Coefficients outside parentheses depict effects after controlling for perceptions of partners’ destructive behavior. *p < .05; **p < .01; ***p < .001.

Perceivers’ negative emotion was regressed on target partners’ destructive conflict behavior and perceivers’ valuing of target partners, controlling for gender and relationship length. Perceivers experienced more negative emotion when target partners enacted more destructive conflict behavior (Path D), b = .61, β = .34, t = 5.19, p < .001. In addition, perceivers experienced less negative emotion when they highly valued target partners (Path E), b = −.24, β = −.16, t = −2.58, p < .05. Neither gender nor relationship length moderated the effects of target partners’ enacted destructive behavior, ps > .58, nor partner valuing, ps > .06, on perceivers’ negative emotion.

Finally, perceivers’ perceptions of the target partner’s destructive conflict behavior were included as an additional predictor. Perceivers who viewed the target partners’ conflict behavior as destructive experienced more negative emotion following the interaction (Path C), b = .57, β = .49, t = 5.86, p < .001. Gender moderated the effect of perceivers’ perceptions on experienced negative emotion, b = .40, β = .17, t = 1.98, p < .05, such that the effect of perceivers’ perceptions on experienced negative emotion was stronger for females, b = .83, t = 5.45, p < .001, than for males, b = .43, t = 3.37, p < .01. Relationship length did not moderate the effect of perceivers’ perceptions, p = .33. In addition, the other effects on perceivers’ emotion were no longer significant once perceivers’ perceptions of targets’ destructive behavior were controlled, b = −.02, β = −.01, t = −.16, p = .88 (Path D’); and b = −.14, β = −.09, t = −1.52, p = .13 (Path E’). The indirect effects were tested using Monte Carlo tests (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002; Selig & Preacher, 2008). These analyses were used to generate 95% confidence intervals (CIs) based on a simulation of 20,000 resamples of the data. The indirect effect of partners’ destructive behavior on perceivers’ negative emotion via perceptions of targets’ destructive behavior (Path AxC) was significant, 95% CI [.41, .86], suggesting that accurate detection of target partners’ destructive conflict behavior was a significant predictor of perceivers’ negative emotion. The indirect effect of perceivers’ valuing of partners on perceivers’ emotion via perceivers’ perceptions of the target partner’s destructive behavior (Path BxC) was also significant, 95% CI [−.17, −.03], suggesting that biased misperceptions of the target partner’s destructive conflict behavior also significantly predicted perceivers’ negative emotion.

Additional analyses

Additional analyses were conducted to test alternative explanations and address potential limitations of the analyses described earlier. It is possible that the index of target partners’ actual destructive behavior is flawed because it was an average of coders’ ratings and target partners’ self-reports. Perhaps evidence for accuracy would be greater (and perhaps bias effects would not be found) if we instead modeled these two components as separate predictors. However, in an additional analysis, we found that perceivers’ valuing of target partners predicted their perceptions of target partners’ destructive conflict behavior even when including these two accuracy benchmarks as separate predictors, b = −.16, p < .01. Coders’ ratings of target partners’ destructive conflict behavior and target partners’ self-reports of their own destructive conflict behaviors independently predicted perceivers’ judgments of target partners’ destructive behavior, b = .68 and .43, respectively, ps < .001, suggesting that each component of the index captures a somewhat unique aspect of the target partner’s behavior despite their high correlation.

Further, since people project their own behaviors onto their perceptions of their partner (e.g., Lemay et al., 2007; Murray et al., 1996b), it might be the case that perceivers’ own behavior plays an important role in their perceptions of their partner’s conflict behavior. More specifically, it might be the case that when perceivers highly value their partners, they behave less destructively, and in turn, they may project their own less destructive behavior onto their partner. Therefore, we created an index of perceivers’ own destructive behavior by averaging coders’ ratings of perceivers’ behavior and perceivers’ ratings of their own behavior. We then examined whether perceivers’ behavior served as a mediator in explaining why perceivers’ valuing of their partners biases their perceptions of their partner’s destructive behaviors, with gender and relationship length included as covariates in all analyses. Participants who valued their partners enacted less destructive conflict behavior, b = −.21, t = −4.17, p < .001. In turn, controlling for accuracy of partner’s behavior, we found that perceivers’ own destructive behavior predicted their perceptions of their partner’s destructive behavior, b = .48, t = 5.94, p < .001, and we continued to find evidence for accuracy, b = .80, t = 10.16, p < .001, and bias via perceivers’ value for their partner became marginal, b = −.10, t = −1.80, p = .07. The indirect effect of perceivers’ partner valuing on their perceptions of their partner’s behavior via their own behavior was significant, 95% CI (−.16, −.05), suggesting that projection processes may be partially explaining why partner valuing biases partner perceptions. Further, we continued to find that even after adding perceivers’ behavior as a covariate, perceivers who viewed their target partners’ conflict behavior as destructive experienced more negative emotion following the interaction, b = .42, t = 4.11, p < .001.

It is possible that perceivers’ confidence in being valued by partners, rather than their own valuing of partners, biases their judgments of their partner’s destructive conflict behavior. In an additional analysis, perceivers’ security in their partner’s care, regard, and commitment was added as an additional predictor. This analysis continued to find evidence for accuracy, b = 1.09, p < .001, though bias via perceivers’ valuing of their partner became marginally significant, b = −.16, p = .06. Perceivers’ security did not predict perceivers’ perceptions of their partner’s destructive behavior, p = .91, suggesting that security (i.e., trust in the partner’s care, regard, and commitment) could not explain the current bias effects.

Some readers may be curious regarding whether each component of the index of perceivers’ valuing of partners predicts perceivers’ judgments of their partner’s destructive conflict behavior. This was found to be the case. In separate analyses, we found that perceivers’ perceptions of their partner’s destructive behavior were marginally or significantly predicted by their commitment, b = −.07, p = .09, their care for their partner, b = −.14, p < .01, and their global regard for their partner, b = −.17, p < .01.

It is possible that the outcome variable (negative emotion) and the putative mediator variable (perceptions of partner’s destructive conflict behavior) are so strongly related that either variable could serve as the mediator. That is, perhaps negative emotion mediates the effects of the predictors on perceptions of the partner’s behavior. This was not the case. Support for Paths A and B (described earlier) was found even in a model that controlled for negative emotion, ps < .05. Hence, full mediation was found in a model that tested perceptions of destructive conflict behavior as a mediator predicting emotion, but not in an alternative model that examined emotion as a mediator predicting perceptions of destructive conflict behavior.

Discussion

This study supported predictions regarding accuracy and bias in perceptions of romantic partners’ destructive conflict behavior. Perceivers’ judgments of the target partner’s behavior were predicted by the index of the target partner’s actual behavior (an average of coders’ judgments and the target partners’ self-reports of their own behavior). Independent of this accuracy effect, perceivers’ judgments were also predicted by their chronic valuing of the target partner. That is, perceivers who strongly valued a close relationship with their partner underestimated the destructive nature of their partner’s behavior during the conflict discussion, whereas perceivers who had relatively more negative sentiments toward their partner overestimated their partner’s destructive conflict behavior. In turn, participants who perceived their partner as engaging in destructive behavior experienced more negative emotion during the interaction, and both accuracy and bias processes contributed to this effect. Our interpretation of these findings will be discussed in turn.

Accurate and biased perceptions of partner’s destructive conflict behavior

Our finding that people are able to detect their partner’s destructive behavior with some accuracy is consistent with numerous findings that individuals are able to accurately identify their partner’s behaviors and feelings (e.g., Collins & Feeney, 2000; Coriell & Cohen, 1995; Fletcher & Kerr, 2010; Lemay & Neal, 2013; Lemay et al., 2007), as well as their partner’s generalized conflict styles (Segrin, Hanzal, & Domschke, 2009). Notably, we found support for accuracy both when the reality of the partner’s destructive behavior was computed as the average of the partner’s self-report and the coders’ objective ratings, as well as when these two components were specified as separate predictors.

In addition to accuracy, we found that perceptions of partner’s destructive conflict behaviors were derived from bias. This is consistent with prior research that suggests that people who strongly value their relationships want to view their romantic partners as caring, committed, and responsive (Clark & Lemay, 2010; Lemay et al., 2007; Lemay & Melville, 2014; Lemay & Neal, 2013), and that motivation to reach a certain conclusion leads people to utilize specific beliefs and strategies that will make such a conclusion possible (Kunda, 1990). Notably, we found that when individuals are faced with the threat of their valued partner’s destructive behavior during an unresolved conflict discussion, they underestimated the destructive nature of their partner’s behavior. This finding supports existing literature on how desire to maintain a relationship faciliates positive perceptions in the face of relationship threats (Lemay et al., 2015; Rusbult et al., 2000) and contributes to new understanding of how relationship desires operate to defend relationships during conflict. Further, we found that people’s value for their partner leads them to be less destructive during conflict, which in turn leads them to see their partner’s behavior as less destructive, possibly because they are projecting their perceptions of themselves onto their partner. This offers evidence for the role of projection as a partial mechanism by which partner valuing facilitates positive partner perceptions.

Beyond finding that people who value their partners (and thus want to view their partners positively) have positive perceptions of their partner’s behavior, we also found evidence that the reverse is true as well. That is, individuals who attached relatively lower value to their romantic partners (i.e., 1 SD below the mean) overestimated their partner’s destructive behavior, which suggests that when individuals were relatively lower in commitment, care, and regard for their romantic partner, they utilized strategies that facilitated negative perceptions of their partner (i.e. by interpreting these less valued partners as being more destructive). Perhaps these individuals exaggerated the destructive nature of their partner’s behavior as a way of making the partner less appealing, which would faciliate their desire to distance themselves from the partner, or to justify their negative sentiments about their partner. This is similar to Sillars, Roberts, Leonard, and Dun’s (2000) finding that negative attributions toward romantic partners are more typical of people in dissatisfied relationships. Further, the fact that these people experience more anger and frustration during conflict discussions (Sillars, Roberts, Leonard, & Dun, 2000) supports our finding that biased partner perceptions during a conflict have implications for the perceivers’ felt emotions during the conflict. However, it should be noted that, across our sample, partners were relatively highly valued, so 1 SD below the mean partner valuing score was still a relatively high value score (see Table 1). Lower partner valuing may produce stronger underestimation effects.

Emotional influence of accurate and biased perceptions of destructive behavior

The current results included the somewhat counterintuitive finding that people who highly value their romantic partner experience fewer negative emotions during a relationship conflict with that partner. This seems counterintuitive because it might be expected that fighting with a deeply loved relationship partner, rather than someone less important, would be more unpleasant, especially if the conflict is about the most significant unresolved problem in the relationship. Indeed, prior research suggests that people often feel more hurt when they are mistreated by a highly valued partner (Lemay et al., 2012). However, the current research suggests that less negative emotions are experienced during conflicts with valued partners. This effect was explained by biased perceptions of the partner’s behavior. Perceivers who valued their partners experienced less negative emotions during the conflict because of their tendency to see their partner’s behavior as less destructive. Further, this effect was independent of observers’ reports of the partner’s behavior and the partner’s self-reports of their own behavior. This finding supports previous research that positively biased perceptions of romantic partners facilitate improved emotional experience (Lemay & Neal, 2013, 2014). While past research has found that positively biased partner perceptions are associated with marital satisfaction within the domain of relationship conflict (e.g., Segrin et al., 2009), our findings suggest a new mechanism through which romantic partners remain emotionally resilient in the face of relationship conflict.

Furthermore, we found evidence that accurate detection of partner’s destructive behavior predicted more negative emotion. The effect of the partner’s actual destructive behavior on perceivers’ negative emotion was mediated by perceivers’ perceptions of their partner’s destructive behavior. In other words, people experienced more negative emotion when they accurately detected the destructive nature of their partner’s behavior during the conflict, which is consistent with prior findings that accurately identifying a partner’s negative perceptions of one’s relationship predicts depressive moods (Overall & Hammond, 2013) and conversely that accurate detection of a partner’s responsiveness predicts reduced negative affect (Lemay & Neal, 2014). Taken together, these results suggest that both accurate detection and biased perceptions of partner’s behavior contribute to emotional experience during relationship conflict.

An important consideration is that the couples had their conflict discussions in an unfamiliar, public laboratory setting, which may have led the couples to feel more inclined to be polite and downplay any hostility they felt toward one another (Gottman, 1979). However, this would have restricted the variability in conflict behavior, which would have only worked against our ability to support our predictions, given that restricted range usually reduces associations among variables. Further, having romantic couples discuss the most significant, unresolved problem in their relationship added to the realism of our study design.

Another important consideration is that there is no perfect assessment strategy to measure partner’s actual destructive behavior. Objective coders might misinterpret target’s idiosyncrasies, targets might be self-serving when reporting their own negative behaviors, and as we found in the present research, people who are familiar with the target may be biased by their sentiments toward the target. To combat these limitations, we went beyond most studies’ single-measure approach to accuracy by following Funder’s (1995) recommendation to utilize multiple accuracy benchmarks (i.e., coders’ ratings and target’s ratings), thus allowing us to more comprehensively index reality. Still, given that each measure that went into our index of the partner’s behavior is flawed in its own way, the most we can conclude from our findings is that people who value their partner see their partner’s behavior as less destructive even when this is not perceived by observers and the partners themselves, and that this has implications for their emotional experience.

While the presented model expands understanding of how accurate and biased perceptions can help reduce negative affect in romantic relationships, the nature of the study’s cross-sectional, correlational design does not allow for causal conclusions. However, we found more support for the predicted model relative to a reversed mediation model in which negative affect was treated as the mediator and perceptions of destructive conflict behavior was treated as the outcome. To more directly test the causal effects implied by our model, future experimental research should manipulate participants’ desires to maintain close relationships and examine effects on perceptions of destructive behavior.

Future directions and implications

It may be the case that positive perceptions of a partner’s conflict behaviors are facilitated by an emotion regulation strategy. One such strategy, cognitive reappraisal, involves thinking about situations in ways that control the experience of emotion (Gross, 2002; Gross & John, 2003). For instance, those who have been instructed to utilize reappraisal by thinking of positive aspects regarding their romantic relationship and partner prior to a conflict discussion express less contempt during the conflict than those who are not instructed to do so (Ben-Naim, Hirschberger, Ein-Dor, & Mikulincer, 2013), and this strategy is positively associated with increased psychological well-being, having closer relationships, and being liked by peers (Gross & John, 2003). In light of our findings, it might be the case that participants who valued their partner utilized cognitive reappraisal to facilitate positive perceptions of their partner and thus protected their emotional well-being. Future research should examine if positive perceptions in valued relationships are in fact facilitated by emotion regulation strategies.

It should also be noted that it may not always be advantageous to underestimate a partner’s destructive behavior, as there may be certain situations in which such perceptions are detrimental to partner and relationship well-being. For example, objectively harmful (but subjectively valued) relationships may be prolonged if people consistently underestimate their partner’s repeated destructive behaviors. It may also be the case that if an individual consistently sees his or her partner as being less destructive than that partner sees him or herself, this over-idealization may also be detrimental to the welfare of the relationship (Tomlinson, Aron, Carmichael, Reis, & Holmes, 2014). For example, underestimating how upset a romantic partner is during an important conflict may interfere with successful conflict resolution. Therefore, though underestimating partner’s destructive behaviors may temporarily improve emotional well-being during the conflict, it may also come with costs for the long-term functioning of the relationship.

Summary

In conclusion, we found support for our model of accurate and biased perceptions of partner behavior as determinants of emotional experience during relationship conflict. That is, individuals were both accurate and biased in their perceptions of their partner’s destructive conflict behavior. Positively distorted perceptions served as a mechanism for regulating emotional experience during the conflict. Notably, individuals who attached high value to their romantic partners were able to mitigate the negative emotions they experienced during their relationship conflict not in spite of their positive sentiments toward their partners, but because of them. People who strongly care about their partners seem motivated to view their romantic relationships positively, and the resulting underestimation of their partner’s destructive behavior appears to buffer against the negative emotions they would be vulnerable to if they saw their partner’s behavior with complete accuracy. In consideration of these findings, distorting perceptions of destructive conflict behavior is likely to be just a single application of how value for one’s romantic partner regulates emotional experience and prompts future research in other relationship contexts where value for one’s partner may improve emotional well-being. Indeed, it may be the case that the ignorance reflected in positive relationship illusions does, in fact, usually lead to bliss.

Footnotes

Authors’ note

Other findings from this study have been published elsewhere (Lemay, 2014; Lemay & Dobush, 2015; Lemay & Spongberg, 2015), but these papers addressed different research questions, none on perceptions of hostility. The findings reported in the current manuscript have not been previously reported, though they have been presented at the Annual Meeting of the Association for Psychological Science and an International Association for Relationship Research mini-conference.

Acknowledgment

The authors would like to thank the members of the Interpersonal Relationships Lab at the University of New Hampshire for their assistance with data collection and coding.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Foundation (grant BCS 1145349).

Open research statement

As part of IARR’s encouragement of open research practices, the authors have provided the following information: This research was not pre-registered. The data and materials used in the research are available upon request by emailing Edward Lemay ().

ORCID iD

Rachel B. Venaglia

Note

References

Abbey

Andrews

F. M.

Halman

L. J.

(1995). Provision and receipt of social support and disregard: What is their impact on the marital life quality of infertile and fertile couples? Journal of Personality and Social Psychology, 68, 455–469.

Baron

K. G.

Smith

T. W.

Butner

Nealey-Moore

Hawkins

M. W.

Uchino

B. N.

(2007). Hostility, anger, and marital adjustment: Concurrent and prospective associations with psychosocial vulnerability. Journal of Behavioral Medicine, 30, 1–10.

Ben-Naim

Hirschberger

Ein-Dor

Mikulincer

(2013). An experimental study of emotion regulation during relationship conflict interactions: The moderating role of attachment orientations. Emotion, 13, 506–519.

Berscheid

Ammazzalorso

(2004). Emotional experience in close relationships. In Brewer

M. B.

Hewstone

(Eds.), Emotion and motivation (pp. 47–69). Malden, MA: Blackwell Publishing.

Bradbury

T. N.

Fincham

F. D.

(1990). Attributions in marriage: Review and critique. Psychological Bulletin, 107, 3–33.

Braiker

H. B.

Kelley

H. H.

(1979). Conflict in the development of close relationships. In Burgess

R. L.

Huston

T. L.

(Eds.), Social exchange in developing relationships (pp. 135–168). New York, NY: Academic Press, Inc.

Canary

D. J.

Cupach

W. R.

Messman

S. J.

(1995). Series editors’ introduction. In Hendrick

Hendrick

S. S.

(Eds.), Relationship conflict: Conflict in parent-child, friendship, and romantic relationships (pp. vii–viii). Thousand Oaks, CA: SAGE.

Clark

M. S.

Lemay

E. P.

Jr . (2010). Close relationships. In Fiske

S. T.

Gilbert

D. T.

Lindzey

(Eds.), Handbook of social psychology (5th ed., Vol. 2, pp. 898–940). New York, NY: Wiley.

Collins

N. L.

Feeney

B. C.

(2000). A safe haven: An attachment theory perspective on support seeking and caregiving in intimate relationships. Journal of Personality and Social Psychology, 78, 1053–1073.

10.

Coriell

Cohen

(1995). Concordance in the face of a stressful event: When do members of a dyad agree that one person supported the other? Journal of Personality and Social Psychology, 69, 289–299.

11.

Downey

Freitas

A. L.

Michaelis

Khouri

(1998). The self-fulfilling prophecy in close relationships: Rejection sensitivity and rejection by romantic partners. Journal of Personality and Social Psychology, 75, 545–560.

12.

Fincham

F. D.

Paleari

F. G.

Regalia

(2002). Forgiveness in marriage: The role of relationship quality, attributions, and empathy. Personal Relationships, 9, 27–37.

13.

Fletcher

G. J.

Kerr

P. S.

(2010). Through the eyes of love: Reality and illusion in intimate relationships. Psychological Bulletin, 136, 627–658.

14.

Funder

D. C.

(1995). On the accuracy of personality judgment: A realistic approach. Psychological Review, 102, 652–670.

15.

Gottman

J. M.

(1979). Marital interaction: Experimental investigations. New York, NY: Academic Press.

16.

Gottman

J. M.

(1993). The roles of conflict engagement, escalation, and avoidance in marital interaction: A longitudinal view of five types of couples. Journal of Consulting and Clinical Psychology, 61, 6–15.

17.

Gross

J. J.

(2002). Emotion regulation: Affective, cognitive, and social consequences. Psychophysiology, 39, 281–291.

18.

Gross

J. J.

John

O. P.

(2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85, 348–362.

19.

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.

20.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis (methodology in the social sciences). New York, NY: Guilford Press.

21.

Kunda

(1990). The case for motivated reasoning. Psychological Bulletin, 108, 480–498.

22.

Leary

M. R.

Springer

Negel

Ansell

Evans

(1998). The causes, phenomenology, and consequences of hurt feelings. Journal of Personality and Social Psychology, 74, 1225–1237.

23.

Lemay

E. P.

Jr . (2014). Accuracy and bias in self-perceptions of responsive behavior: Implications for security in romantic relationships. Journal of Personality and Social Psychology, 107, 638–656.

24.

Lemay

E. P.

Jr 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.

25.

Lemay

E. P.

Jr Clark

M. S.

(2015). Motivated cognition in relationships. Current Opinion in Psychology, 1, 72–75.

26.

Lemay

E. P.

Jr Clark

M. S.

Feeney

B. C.

(2007). Projection of responsiveness to needs and the construction of satisfying communal relationships. Journal of Personality and Social Psychology, 92, 834–853.

27.

Lemay

E. P.

Jr Lin

J. L.

Muir

H. J.

(2015). Daily affective and behavioral forecasts in romantic relationships: Seeing tomorrow through the lens of today. Personality and Social Psychology Bulletin, 41, 1005–1019.

28.

Lemay

E. P.

Jr Melville

M. C.

(2014). Diminishing self-disclosure to maintain security in partners’ care. Journal of Personality and Social Psychology, 106, 37–57.

29.

Lemay

E. P.

Jr Neal

A. M.

(2013). The wishful memory of interpersonal responsiveness. Journal of Personality and Social Psychology, 104, 653–672.

30.

Lemay

E. P.

Jr Neal

A. M.

(2014). Accurate and biased perceptions of responsive support predict well-being. Motivation and Emotion, 38, 270–286.

31.

Lemay

E. P.

Jr Overall

N. C.

Clark

M. S.

(2012). Experiences and interpersonal consequences of hurt feelings and anger. Journal of Personality and Social Psychology, 103, 982–1006.

32.

Lemay

E. P.

Jr Dobush

(2015). When do personality and emotion predict destructive behavior during relationship conflict? The role of perceived commitment asymmetry. Journal of Personality, 83, 523–534.

33.

Lemay

E. P.

Jr Pruchno

R. A.

Feild

(2006). Accuracy and bias in perceptions of spouses’ life-sustaining medical treatment preferences. Journal of Applied Social Psychology, 36, 2337–2361.

34.

Lemay

E. P.

Jr Spongberg

(2015). Perceiving and wanting to be valued by others: Implications for cognition, motivation, and behavior in romantic relationships. Journal of Personality, 83, 464–478.

35.

MacKenzie

Smith

T. W.

Uchino

White

P. H.

Light

K. C.

Grewen

K. M.

(2014). Depressive symptoms, anger/hostility, and relationship quality in young couples. Journal of Social and Clinical Psychology, 33, 380–396.

36.

MacKinnon

D. P.

Lockwood

C. M.

Hoffman

J. M.

West

S. G.

Sheets

(2002). A comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7, 83–104.

37.

McGonagle

K. A.

Kessler

R. C.

Schilling

E. A.

(1992). The frequency and determinants of marital disagreements in a community sample. Journal of Social and Personal Relationships, 9, 507–524.

38.

Mills

Clark

M. S.

Ford

T. E.

Johnson

(2004). Measurement of communal strength. Personal Relationships, 11, 213–230.

39.

Murray

S. L.

Bellavia

G. M.

Rose

Griffin

D. W.

(2003). Once hurt, twice hurtful: How perceived regard regulates daily marital interactions. Journal of Personality and Social Psychology, 84, 126–147.

40.

Murray

S. L.

Holmes

J. G.

Griffin

D. W.

(1996a). The benefits of positive illusions: Idealization and the construction of satisfaction in close relationships. Journal of Personality and Social Psychology, 70, 79–98.

41.

Murray

S. L.

Holmes

J. G.

Griffin

D. W.

(1996b). The self-fulfilling nature of positive illusions in romantic relationships: Love is not blind, but prescient. Journal of Personality and Social Psychology, 71, 1155–1180.

42.

Murray

S. L.

Holmes

J. G.

Griffin

D. W.

(2000). Self-esteem and the quest for felt security: How perceived regard regulates attachment processes. Journal of Personality and Social Psychology, 78, 478–498.

43.

Overall

N. C.

Fletcher

G. J.

Simpson

J. A.

Sibley

C. G.

(2009). Regulating partners in intimate relationships: The costs and benefits of different communication strategies. Journal of Personality and Social Psychology, 96, 620–639.

44.

Overall

N. C.

Hammond

M. D.

(2013). Biased and accurate: Depressive symptoms and daily perceptions within intimate relationships. Personality and Social Psychology Bulletin, 39, 636–650.

45.

Peterson

J. L.

DeHart

(2014). In defense of self-love: An observational study on narcissists’ negative behavior during romantic relationship conflict. Self and Identity, 13, 477–490.

46.

Roseman

I. J.

Spindel

M. S.

Jose

P. E.

(1990). Appraisals of emotion-eliciting events: Testing a theory of discrete emotions. Journal of Personality and Social Psychology, 59, 899–915.

47.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

48.

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.

49.

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.

50.

Rusbult

C. E.

Van Lange

P. A.

Wildschut

Yovetich

N. A.

Verette

(2000). Perceived superiority in close relationships: Why it exists and persists. Journal of Personality and Social Psychology, 79, 521–545.

51.

Scherbaum

C. A.

Ferreter

J. M.

(2009). Estimating statistical power and required sample sizes for organizational research using multilevel modeling. Organizational Research Methods, 12, 347–367.

52.

Segrin

Hanzal

Domschke

T. J.

(2009). Accuracy and bias in newlywed couples’ perceptions of conflict styles and the association with marital satisfaction. Communication Monographs, 76, 207–233.

53.

Selig

J. P.

Preacher

K. J.

(2008). Monte Carlo method for assessing mediation: An interactive tool for creating confidence intervals for indirect effects [Computer software]. Retrieved June 2008, from http://quantpsy.org/

54.

Sillars

Roberts

L. J.

Leonard

K. E.

Dun

(2000). Cognition during marital conflict: The relationship of thought and talk. Journal of Social and Personal Relationships, 17, 479–502.

55.

Simpson

J. A.

Ickes

Blackstone

(1995). When the head protects the heart: Empathic accuracy in dating relationships. Journal of Personality and Social Psychology, 69, 629–641.

56.

Simpson

J. A.

Rholes

W. S.

Phillips

(1996). Conflict in close relationships: An attachment perspective. Journal of Personality and Social Psychology, 71, 899–914.

57.

Simpson

J. A.

Rholes

W. S.

Winterheld

H. A.

(2010). Atachment working models twist memories of relationship events. Psychological Science, 21, 252–259.

58.

Tomlinson

J. M.

Aron

Carmichael

C. L.

Reis

H. T.

Holmes

J. G.

(2014). The costs of being put on a pedestal: Effects of feeling over-idealized. Journal of Social and Personal Relationships, 31, 384–409.

59.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.