Is Empathic Accuracy Enough to Facilitate Responsive Behavior in Dyadic Interaction? Distinguishing Ability From Motivation

Abstract

Growing evidence suggests that interpersonal responsiveness—feeling understood, validated, and cared for by other people—plays a key role in shaping the quality of one’s social interactions and relationships. But what enables people to be interpersonally responsive to others? In the current study, we argued that responsiveness requires not only accurate understanding but also compassionate motivation. Specifically, we hypothesized that understanding another person’s thoughts and feelings (empathic accuracy) would foster responsive behavior only when paired with benevolent motivation (empathic concern). To test this idea, we asked couples (N = 91) to discuss a personal or relationship stressor; we then assessed empathic accuracy, empathic concern, and responsive behavior. As predicted, when listeners’ empathic concern was high, empathic accuracy facilitated responsiveness; but when empathic concern was low, empathic accuracy was unhelpful (and possibly harmful) for responsiveness. These findings provide the first evidence that cognitive and affective forms of empathy work together to facilitate responsive behavior.

Keywords

empathic accuracy empathic concern responsiveness empathy understanding open data open materials

When people disclose important personal concerns to others, they often assume that listeners will be more caring and responsive when they accurately understand the disclosers’ thoughts, feelings, and needs. This ability to accurately infer another person’s internal state is known as empathic accuracy (Ickes, 1993, 1997). Because this ability is assumed to be fundamental to healthy social development and effective navigation of the social world, most of the research on empathic accuracy has focused on factors that facilitate or impair it (e.g., Davis & Kraus, 1997; Ickes, Gesn, & Graham, 2000; Simpson et al., 2011) and on its underlying neural bases (e.g., Decety & Ickes, 2009; Zaki, Weber, Bolger, & Ochsner, 2009). Very few studies have examined the downstream effects of empathic accuracy on actual social behavior. As a result, there is surprisingly little evidence testing the widely held assumption that empathic accuracy fosters empathic concern (compassion and sensitivity toward other people). Is it enough to simply understand another person’s thoughts and feelings, or is the story more complicated than that?

In the current study, we investigated the distinct roles of empathic accuracy and empathic concern in shaping responsive behavior in couples. Drawing from the broader literature on empathy, we conceptualized empathic accuracy as a cognitive form of empathy that is distinct from the affective form of empathy (empathic concern) that motivates prosocial behavior (Batson, 2009). We suggest that empathic accuracy provides knowledge about how to be responsive to other people, whereas empathic concern motivates people to want to be responsive.

Empathic Accuracy and Responsiveness

Interpersonal responsiveness plays a vital role in social interaction. Responsiveness describes how interaction partners attend to and support one another’s needs, goals, concerns, and perspectives (Reis & Clark, 2013; Reis & Gable, 2015). The dominant theoretical model posits three components of responsiveness: (a) understanding, or accurate insight into a partner’s core self, including his or her needs, desires, concerns, perspectives, or traits; (b) validation, or respect for and valuing of the partner’s core self; and (c) caring, or expressions of affection and concern for the partner’s well-being (Reis & Clark, 2013; Reis & Gable, 2015; Reis & Shaver, 1988). Responsiveness is crucial to any interaction in which partners disclose significant personal information, such as confiding about personal stressors, negotiating conflict, sharing good news, or discussing hopes and dreams for the future.

There is growing evidence that responsiveness shapes the quality of social interactions and relationships. For example, perceived responsiveness enhances the benefits of social support (e.g., Collins & Feeney, 2000; Maisel & Gable, 2009) and fosters trust and intimacy in romantic couples (Gable, Gosnell, Maisel, & Strachman, 2012; Laurenceau, Barrett, & Rovine, 2005; Reis, 2014) and friendships (Canevello & Crocker, 2010; Crocker & Canevello, 2008). Most of this research, however, has focused on the consequences of perceived responsiveness rather than on the factors that predict responsive behavior toward other people (Forest, Kille, Wood, & Holmes, 2014). If perceived responsiveness is critical for well-functioning relationships, what enables people to be responsive to others?

Although theoretical models and intuition suggest that empathic accuracy should facilitate responsive behavior (Reis & Patrick, 1996), only two studies have examined the link between empathic accuracy and outcomes related to responsiveness. In one study, empathic accuracy was assessed after married couples discussed one partner’s personal stressor (Verhofstadt, Buysse, Ickes, Davis, & Devoldre, 2008). Spouses with higher empathic accuracy provided more instrumental (but not emotional) support and less negative support. In a second study, newlyweds with higher empathic accuracy during laboratory conflict discussions reported more accommodation during conflicts at home, which suggests that empathic accuracy may help prevent conflicts from escalating (Kilpatrick, Bissonnette, & Rusbult, 2002).

These studies provide preliminary evidence that empathic accuracy may facilitate responsive behavior in social support and conflict interactions. But does accurate understanding always foster prosocial action? After all, individuals may know exactly what their partner is thinking and feeling, but at the same time feel annoyed, frustrated, or burdened instead of compassionate and caring. Likewise, accuracy may reveal irreconcilable differences that undermine closeness (Sillars, 1985; Simpson, Oriña, & Ickes, 2003) or increase anxious hypervigilance to relationship threats (Simpson, Ickes, & Blackstone, 1995). Even competitors or enemies try to understand each other, but their insights likely promote self-serving behaviors rather than altruistic ones (Gilin, Maddux, Carpenter, & Galinsky, 2013).

From this perspective, there is no reason to assume that empathic accuracy is itself benevolent (Ickes, in press). Instead, we conceptualize empathic accuracy as a perceptual skill from which people glean important insights about the inner states of others. These insights can be used to promote the best interests of others or to fulfill more egocentric motives. Thus, empathic accuracy has the potential—but is not likely to be sufficient—to facilitate responsive behavior. We suggest that effects of empathic accuracy will depend on the listener’s degree of prosocial motivation. Indeed, responsiveness theory assumes that listeners must first intend to understand, validate, and care for the needs of the discloser (Reis & Gable, 2015). We argue that this intention to care for the discloser, and to serve his or her best interests, is at the heart of determining whether empathic accuracy will foster responsiveness.

The Moderating Role of Empathic Motivation

Research on prosocial behavior distinguishes cognitive empathy (empathic accuracy, perspective taking) from its affective counterpart—empathic concern. Empathic concern refers to in-the-moment feelings of warmth, compassion, and tenderness that motivate prosocial behavior among strangers, even at the expense of one’s own time or resources (Batson, 1991, 2009). The close-relationships literature also recognizes the importance of compassionate motivation in responding to the needs of close others. For example, people with higher levels of compassionate love (Collins et al., 2014; Sprecher & Fehr, 2005), compassionate (vs. self-image) goals (Canevello & Crocker, 2010), and altruistic (vs. egoistic) caregiving motivations (Feeney & Collins, 2003) are more supportive to friends and romantic partners. And when people feel greater empathic concern while watching their romantic partner undergo a stressful task, they are more affectionate and emotionally supportive afterward (Collins et al., 2014).

We suggest that empathic concern motivates caring behavior, but empathic accuracy enables listeners to be authentically understanding and validating. In the absence of accurate insight, even well-intentioned listeners may be handicapped in their ability to be optimally responsive. In such cases, expressions of validation and caring may be perceived as insincere, insensitive, or out of sync with disclosers’ goals and needs (Collins, Ford, Guichard, Kane, & Feeney, 2010; Reis & Clark, 2013; Reis & Gable, 2015). Although prior theoretical work implies that cognitive and affective forms of empathy may depend on one another, no researchers to date have explicitly hypothesized about or tested the interactive effects of empathic accuracy and empathic concern on social behavior.

The Current Study

The primary goal of this study was to investigate the joint effects of empathic accuracy and empathic concern on responsive behavior. We brought couples into the lab and asked them to discuss one partner’s (the target’s) relationship or personal stressor. We then measured the other partner’s (the perceiver’s) empathic accuracy, empathic concern, and responsive behavior. Responsive behavior can be measured in a variety of ways, ranging from microlevel coding of paralinguistic and nonverbal behaviors (e.g., turn taking, gaze, gesticulation; Duncan & Fiske, 1979) to more macrolevel coding of expressed understanding, caring, and validation (see Maisel, Gable, & Strachman, 2008). In keeping with the interpersonal-process model underlying responsiveness theory (Reis & Shaver, 1988), we limited our focus to macrolevel, or global, perceptions of enacted responsiveness—though we acknowledge that targets (and observers) likely incorporate microlevel cues when making macrolevel judgments (see Reis & Clark, 2013; Reis, Clark, & Holmes, 2004; Reis & Shaver, 1988). In addition, to assess both the objective and subjective qualities of the interaction, we asked targets (the discloser or support recipient) and trained observers to rate the perceiver’s responsiveness. These ratings were then modeled as a latent variable to maximize the reliability and validity of our behavioral assessment (Collins & Feeney, 2000).

Our conceptual model is depicted in Figure 1. We hypothesized that the effect of perceivers’ empathic accuracy on their responsive behavior would be moderated by their empathic concern. Specifically, when perceivers’ empathic concern is high, empathic accuracy should facilitate responsive behavior. However, when perceivers’ empathic concern is low, empathic accuracy should fail to facilitate (and may even hinder) responsiveness.

Fig. 1.

Conceptual model tested in the present study. We hypothesized that the effect of perceivers’ empathic accuracy (accurate understanding of targets’ thoughts and feelings) on their responsive behavior toward targets would be moderated by perceivers’ empathic concern (caring motivation). Perceivers’ responsive behavior was measured as a latent variable (indicated by the oval) that was composed of target and observer ratings. The rectangles indicate observed variables.

A secondary goal was to explore whether the effects of empathic accuracy and empathic concern depend on the conversation topic (relational vs. personal stressor). Although we had no theoretical reason to expect differences as a function of the discussion topic, we examined both types of interactions to test whether these processes generalized across interpersonal contexts.

Method

Participants

Participants were 91 romantic couples (86 heterosexual, 5 same sex) recruited from the University of California, Santa Barbara, undergraduate participant pool and campus community. Five additional couples were recruited for the study but were not included in the analysis: For 3 of these couples, malfunctioning video equipment prevented us from collecting empathic-accuracy or responsiveness data, and the other 2 did not follow instructions during the videotaped discussion. On the basis of prior work in our lab, we aimed to recruit 100 couples over the course of one academic year and stopped at the end of the academic year when the participant pool and financial resources were exhausted. Couples were required to be in a romantic relationship for at least 3 months. The average relationship length was 14.42 months (SD = 13.63), and 20% of couples were cohabiting. Participants’ mean age was 19.32 years (SD = 1.32). Participants classified themselves as White (50%), Hispanic (19%), Asian/Pacific Islander (20%), African American (3%), American Indian (1%), and “other” (7%). Couple members received either course credit or a $5 gift card for their participation.

Procedure

Couples were recruited for a study about communication skills in romantic relationships. Before the laboratory session, each partner completed an online questionnaire at home assessing a variety of personality and relationship variables. For the current investigation, we used only the demographic items and a measure of relationship satisfaction. Relationship satisfaction was measured with the 7-item Relationship Assessment Scale (Hendrick, 1988; α = .79). Sample items include “In general, how satisfied are you with your relationship?” (1, unsatisfied, to 5, extremely satisfied) and “How many problems are there in your relationship?” (1, very few, to 5, very many; reverse-coded).

Approximately 1 week later, couples arrived at the laboratory and were informed that they would be videotaped discussing topics that might include areas of conflict, personal stressors, and positive relationship events. We randomly assigned one partner (39 men, 52 women) to be the perceiver (the support provider) and the other partner (49 men, 42 women) to be the target (the support recipient). Partners were led to separate rooms to complete a brief preinteraction questionnaire that would be used to choose the target’s topic for discussion.¹ Both partners were asked to list and rate the importance of three areas of conflict in their relationship as well as a salient personal stressor (one that was unrelated to their relationship) that they would be willing to discuss with their partner in the lab. Importance ratings were made on a scale from 1 (not at all important) to 7 (extremely).

Half of the couples were randomly assigned to the support condition and were asked to discuss the target’s personal stressor (e.g., falling behind in school, work pressures, problems with family or roommate). The other half of the couples were assigned to the conflict condition and asked to discuss one of the conflict topics they had listed (e.g., not spending enough time together, lack of communication, feeling jealous). The experimenter was careful to choose a conflict topic that both partners agreed to discuss. In most cases, targets listed at least one conflict that perceivers had also listed. If partners did not agree on a conflict topic, the experimenter always chose a conflict topic that the target had listed. Across conditions, targets rated their discussion topic as moderately to extremely important (M = 5.24, SD = 1.53).

Interaction task

Participants were reunited in a comfortable room that was furnished to look like a living room. The experimenter informed participants of their conversation topic and confirmed that they would be comfortable discussing the topic. Participants were asked to discuss the issue as they naturally would outside of the lab setting and to stay on topic for the full 5-min interaction. A research assistant stood by in an observation room and began video-recording the couple as soon as the experimenter left the room.

After the 5 min had elapsed, participants were separated into different rooms, where they completed a brief questionnaire assessing their thoughts and feelings during the interaction. Our goal was to assess immediate emotional responses to the interaction—including feelings of empathic concern and responsiveness. However, we kept this questionnaire very brief because we did not want to prime particular ideas prior to the upcoming video-review task. For the current investigation, we focused only on the questionnaire items relevant to our hypotheses.

Video-review procedure

To assess empathic accuracy, we asked participants to complete a validated video-review task (see Ickes, Robertson, Tooke, & Teng, 1986). Participants were told that they would be reviewing a recording of their interaction and that each time they remembered having a thought or feeling during the interaction, they should press “pause,” record the time on the video, and write down the specific content of the thought or feeling they recalled having (using complete sentences). Each partner completed this task in private.

After both partners had listed their own thoughts and feelings, each was asked to watch the video again and infer the other’s thoughts and feelings. Although both participants completed this second viewing, the current article focuses only on the participant in the role of perceiver. The experimenter replayed the video from the beginning, pausing the tape at the precise moments that the target remembered having a thought or feeling and instructing the perceiver to infer those thoughts and feelings. Targets reported remembering an average of 7.62 thoughts and feelings during the first video-review task (SD = 3.56), which meant that perceivers had this many opportunities, on average, to make inferences about what the target was thinking or feeling (range = 3–21).

After the video-review procedure, couple members completed a final questionnaire that assessed their thoughts and feelings about the video-review task and about their interaction with their partner (including additional items designed to assess perceived partner responsiveness). We administered this questionnaire after the video-review procedure to avoid contaminating participants’ memory or construal of their interaction. For the present study, we focused only on the items assessing targets’ perceptions of their partner’s responsiveness toward them during their interaction.

Positive-interaction task and debriefing

To offset any discomfort resulting from the conflict or support discussion, we brought couples back together afterward and asked them to spend a few minutes discussing a memory of a positive relationship event. Once they had chosen the topic, the experimenter left the room and allowed participants to converse for 3 min. Participants were then carefully debriefed and thanked for their time.

Predictor variables

Perceivers’ empathic concern

Immediately after their interaction and prior to the video-review task, perceivers rated the degree to which they experienced empathic concern for their partner during the interaction using four items adapted from prior research (Batson, Early, & Salvarani, 1997). Specifically, perceivers rated the degree to which they felt “warm toward my partner,” “compassionate toward my partner,” “close to my partner,” and “sympathetic toward my partner” (α = .87). Items were rated on a 5-point scale (0, not at all, to 4, extremely).

Perceivers’ empathic accuracy

To construct empathic-accuracy scores for perceivers, we carefully transcribed and organized all couple members’ descriptions of their thoughts and feelings in chronological order so that raters could see each actual thought or feeling (from targets) and its corresponding inferred thought or feeling (from perceivers). Using procedures outlined by Stinson and Ickes (1992), four raters (two male, two female) who were blind to condition and study hypotheses rated the degree of similarity between the actual thought or feeling and each corresponding inference by assigning a score of 0 (no match), 1 (similar content but not the same), or 2 (essentially the same content). Interrater reliabilities (αs) ranged from .83 to .96 across codings. We computed empathic-accuracy scores by averaging across the four ratings for each inference and taking the sum of these averages. To adjust for variability in the number of inferences each perceiver had to make, we then divided this sum by the highest number of accuracy points that the perceiver could have earned. The final empathic-accuracy score ranged from 0 to 1, with higher scores indicating greater accuracy. Two empathic-accuracy scores were more than 2 standard deviations above the mean. We Winsorized these values to reduce their influence while preserving their rank order (Tabachnick & Fidell, 2007). As a result, the sample mean changed from .309 (SD = .177) to .306 (SD = .170).

Dependent variable: perceiver’s responsive behavior

To provide the most valid and reliable assessment of the perceiver’s responsive behavior toward the target, we measured responsiveness in two ways. First, we asked targets to report the degree to which their partner (the perceiver) had been supportive and responsive toward them during the interaction. Second, we asked objective raters to view the videotapes and code the quality of each perceiver’s responsive behavior using a validated coding scheme (Maisel et al., 2008). We used these two complementary and correlated sources of information to create a latent variable reflecting the perceiver’s responsiveness toward the target. (For a more detailed discussion of our operationalization of responsiveness, see the Supplemental Material available online.)

Targets’ ratings of perceiver responsiveness

We assessed targets’ ratings of perceiver responsiveness on two occasions, once immediately after the interaction (Time 1) and again after the video-review task (Time 2). The Time 1 items were designed to capture the targets’ most immediate sense of perceiver responsiveness during the interaction. We embedded these items within a larger emotions checklist and kept them brief so that we would not prime any particular thoughts or feelings prior to the video-review task. Targets rated the degree to which their partner was “warm toward me,” “compassionate toward me,” and “close to me” on 5-point scales (0, not at all, to 4, extremely). These items were highly correlated with one another (α = .92) and were therefore averaged to form a composite variable (Time 1 partner responsiveness). This variable was used as the first indicator on the latent dependent variable.

To gain a more complete view of targets’ perceived partner responsiveness, we waited until after the video-review task (Time 2) to measure these perceptions more explicitly. Borrowing from Reis’s (2006) perceived-partner-responsiveness scale, we administered three items that closely mirrored the ratings our coders later used to assess the three components of responsiveness. We asked participants to think back to their interaction and rate (a) the degree to which their partner seemed to understand them, (b) their sense of validation from their partner (e.g., how much their partner showed respect for their opinions and ideas), and (c) the degree to which their partner expressed caring concern for them. Each item was rated on a 7-point scale (1, not at all, to 7, very much). These three items were highly reliable (α = .90) and were averaged to form a composite variable (Time 2 partner responsiveness). This variable was used as the second indicator on the latent dependent variable. (The Time 1 and Time 2 responsiveness variables were highly correlated, r = .72, p < .001.)

Objective ratings of perceiver responsiveness

Using a coding scheme developed by Maisel et al. (2008), three trained raters (one male, two female) who were blind to study hypotheses rated the perceiver’s behavior along three key dimensions of responsiveness: (a) understanding—the degree to which the perceiver listens attentively, gathers information, demonstrates comprehension, and expresses understanding; (b) validation—the extent to which the perceiver shows that he or she values, respects, and accepts the target’s perspectives, concerns, emotions, efforts, or identity and offers reassurance or encouragement; and (c) caring—the degree to which the perceiver expresses love and affection, shows concern, offers support, lifts mood or extends positive mood, and expresses sympathy or empathy. Raters watched each tape twice and then assigned a score for each dimension using a 7-point scale (1, not at all, to 7, very much). Interrater reliabilities (intraclass correlation coefficients, or ICCs) were strong (understanding: ICC = .85; validation: ICC = .80; caring: ICC = .77). Scores for each dimension were then averaged across the raters. These three variables (understanding, validation, and caring) were highly correlated with one another (α = .92) and were therefore averaged to form a composite variable (perceiver responsiveness). This variable was used as the third indicator on the latent dependent variable.

Results

Preliminary analyses

Descriptive statistics and bivariate correlations for all study variables are presented in Table 1. Several patterns are worth noting. First, the correlation between empathic accuracy and empathic concern was not significant (r = .15, p = .146), which indicates that these two forms of empathy are largely independent of one another. Second, perceivers’ relationship satisfaction was unrelated to their empathic-accuracy scores (r = .11, p = .300) but was significantly correlated with their empathic concern (r = .33, p = .001) and all ratings of responsive behavior. Perceivers who were more satisfied with their relationships reported greater empathic concern for their partners and behaved more responsively toward them. Therefore, we controlled for perceiver’s relationship satisfaction in all primary analyses to rule out potential confounds. Finally, we observed a significant gender difference in empathic concern, with male perceivers reporting greater concern (M = 3.31) than female perceivers (M = 2.80), t(89) = 2.83, p = .006. However, there were no gender differences in empathic accuracy or any measure of responsiveness. We also explored whether perceiver gender moderated the effects of empathic accuracy and empathic concern on responsiveness. There were no significant two-way or three-way interactions involving perceiver gender. In addition, controlling for gender did not alter the results reported here. Thus, we collapsed across gender in all analyses.

Table 1.

Descriptive Statistics and Correlations Between Study Variables

Variable	M	1	2	3	4	5	6	7
1. Conversation type (0 = support, 1 = conflict)	.48 (.50)	—
2. Perceivers’ gender (0 = male, 1 = female)	.57 (.50)	−.05	—
3. Perceivers’ relationship satisfaction	4.48 (0.47)	.00	−.07	—
4. Perceivers’ empathic accuracy	.31 (.17)	−.03	−.13	.11	—
5. Perceivers’ empathic concern	3.02 (0.88)	−.33**	−.29*	.33**	.15	—
6. Targets’ ratings of partners’ responsiveness: Time 1 (before video review)	2.72 (1.06)	−.14	−.13	.29*	.16	.45**	—
7. Targets’ ratings of partners’ responsiveness: Time 2 (after video review)	5.72 (1.18)	−.14	−.04	.37**	.08	.46**	.72**	—
8. Observers’ ratings of perceivers’ responsiveness	4.21 (1.39)	−.19^†	−.08	.29*	.18^†	.47**	.45**	.47**

Note: Standard deviations are given in parentheses.

†p < .10. *p < .01. **p < .001.

Primary analyses

To test the joint effects of empathic accuracy and empathic concern on responsive behavior, we conducted latent-variable regression modeling using Mplus (Version 7.0; Muthén & Muthén, 2012) with maximum-likelihood estimation (Kline, 2011). To provide the most valid and reliable assessment of our primary outcome variable—responsive behavior—we began by creating a latent variable comprised of three indicators: (a) target ratings of partner responsiveness at Time 1, (b) target ratings of partner responsiveness at Time 2, and (c) observer ratings of perceivers’ responsiveness. By including both targets’ and observers’ ratings as indicators, we were able to separate out reliable variance (the shared variance between targets and raters) from measurement error unique to each rater. This allowed us to predict responsiveness with greater reliability (and, hence, more statistical power) and validity (in that two people’s agreement on the same interaction was more likely to represent “reality” than the assessment of any one rater alone). In addition, by modeling multiple dependent measures as a latent construct, we were able to conduct a single hypothesis test for a set of correlated variables, thereby reducing the experiment-wise Type I error rate.

To test the adequacy of our latent variable, we conducted a confirmatory factor analysis. The factor loadings indicated that the three responsiveness indicators formed a strong latent factor (see Table 2), which provided evidence of convergent validity. This latent variable captured the shared variance between targets’ ratings of how much their partner (the perceiver) cared for them during the interaction and observers’ ratings of the perceiver’s responsive behavior. We therefore retained the latent variable for the larger regression analysis.

Table 2.

Regression Coefficients (Factor Loadings) From the Latent-Variable Model

Variable	β	b	SE b	z	p
Targets’ ratings of Time 1 partner responsiveness^a	0.84	1.00	—	—	—
Targets’ ratings of Time 2 partner responsiveness	0.84	1.13	0.20	5.73	< .001
Objective ratings of perceiver responsiveness	0.57	0.90	0.18	4.80	< .001

In order to identify the latent variable, we fixed the unstandardized path coefficient for targets’ ratings of Time 1 partner responsiveness at 1.00 and therefore did not test it for significance.

Next, we conducted a latent-variable regression analysis to test our primary hypotheses. In an initial analysis, we predicted responsive behavior (the latent variable) from perceiver empathic accuracy (mean centered), perceiver empathic concern (mean centered), and conversation type (0 = support, 1 = conflict), as well as all two-way interactions and the three-way interaction. We included perceivers’ relationship satisfaction as a control variable. This initial model revealed no significant effects involving conversation type (no main effect, no two-way interactions, and no three-way interaction), which indicates that there were no differences in responsiveness in the two types of conversations, and conversation type did not moderate the effects of empathic concern or empathic accuracy on responsiveness. Therefore, we aggregated effects across the two conversation types and estimated a more parsimonious model with only empathic accuracy, empathic concern, and their two-way interaction. This model fit the data well, χ²(8) = 10.98, p = .203; comparative fit index (CFI) = .98, root-mean-square error of approximation (RMSEA) = .06, standardized root-mean-square residual (SRMR) = .04, and was retained as the final model.

As shown in Table 3, we found no main effect of empathic accuracy (p = .808), which indicates that perceivers who were more accurate were not better (or worse) support providers. There was, however, a strong and significant main effect of empathic concern (p < .001), which indicates that perceivers who experienced more empathic concern during their interaction were more behaviorally responsive to their partners. As predicted, there was also a significant interaction of empathic accuracy and empathic concern (p = .014).

Table 3.

Results of the Latent-Variable Regression Analysis Predicting Responsive Behavior

Predictor	β	b	SE b	95% CI	z	p
Empathic accuracy	0.02	0.12	0.50	[–0.85, 1.09]	0.24	.808
Empathic concern	0.51	0.51	0.11	[0.30, 0.71]	4.84	< .001
Empathic Accuracy ×Empathic Concern	0.23	1.32	0.56	[0.22, 2.41]	2.36	.014
Relationship satisfaction	0.22	0.41	0.18	[0.05, 5.48]	2.23	.025

Note: CI = confidence interval.

To explore this interaction, we computed the simple slope of empathic accuracy as a predictor of responsive behavior at 1 standard deviation above and below the mean of empathic concern (Fig. 2). We found that when empathic concern was high, there was a significant positive association between empathic accuracy and responsive behavior, β = 0.25; b = 1.28, SE = 0.64, 95% confidence interval (CI) = [0.04, 2.53]; z = 2.07, p = .037. Consistent with our prediction, this finding indicates that empathic accuracy facilitated responsive behavior for perceivers whose caring motivation was high. However, when perceivers had low empathic concern, there was a negative association between empathic accuracy and responsive behavior, β = −0.20; b = −1.04, SE = 0.76, 95% CI = [−2.53, 0.44]; z = −1.37, p = .164. Although this trend did not reach significance, it suggests that empathic accuracy was potentially detrimental to responsiveness when perceivers had low caring motivation.

Fig. 2.

Simple slopes predicting responsive behavior from empathic accuracy, separately for individuals with high and low levels of empathic concern. Low and high refer to values 1 standard deviation below and above the mean, respectively.

To further understand this interaction, we computed the simple slope of empathic concern as a predictor of responsiveness for perceivers who had high and low empathic accuracy (±1 SD from the mean). When perceivers had low accuracy, there was a significant positive association between empathic concern and responsiveness (β = 0.29, p = .022). This finding indicates that even when empathic accuracy was relatively low, perceivers with high empathic concern were more responsive to their partners than were perceivers with low empathic concern. However, when empathic accuracy was high, there was a much stronger effect of empathic motivation (β = 0.73, p < .001), which created a much wider gap between perceivers who were high and perceivers who were low in compassionate motivation. This finding indicates that highly compassionate caregivers were even more responsive when they understood their partner, and caregivers with low compassion were even less responsive when they understood their partner. In short, perceivers with high levels of both empathic accuracy and compassionate feelings showed much greater levels of responsiveness than perceivers who were just as accurate about their partners’ thoughts and feelings but felt less compassionate.

Supplementary analyses

The previous analyses suggest that empathic accuracy fosters responsive behavior only when empathic concern is also high. But because perceivers’ empathic concern was moderately correlated with their relationship satisfaction (r = .33, p < .001) and with targets’ relationship satisfaction (r = .37, p < .001), it is possible that the moderating role of empathic concern reflects differences in well- and poorly functioning relationships. That is, to the extent that perceivers in happier relationships tend to feel more empathy toward their partner, perhaps empathic accuracy promotes responsive behavior for happy couples but not for unhappy couples. To test this potential alternative explanation, we reran the moderated latent-variable regression analysis twice: first replacing empathic concern with the perceiver’s relationship satisfaction in the interaction term, then replacing empathic concern with the target’s relationship satisfaction in the interaction term. We found no significant Empathic Accuracy × Relationship Satisfaction interactions predicting responsiveness, which indicates that relationship satisfaction cannot replace empathic concern when explaining the results. We also tested the three-way Empathic Accuracy × Empathic Concern × Relationship Satisfaction interaction and found no significant effects. In other words, the effects of empathic accuracy and empathic concern on responsiveness applied to both satisfied and dissatisfied couples in our sample. (For additional analyses, see the Supplemental Material.)

Discussion

Does empathic accuracy foster interpersonal responsiveness? Although it is often assumed that empathic accuracy promotes positive relationship behavior, few studies have examined the effects of empathic accuracy on actual social behavior. In this study, we examined the role of empathic accuracy and empathic concern in shaping responsive behavior in couples discussing personal or relationship stressors. We focused on responsive behavior as our outcome variable because interpersonal responsiveness is a key predictor of satisfying social interactions and a core feature of secure and well-functioning relationships.

Consistent with our conceptual model, our results showed that the effects of empathic accuracy on responsive behavior depended on compassionate motivation. When empathic concern was high, empathic accuracy was associated with greater responsiveness; when empathic concern was low, empathic accuracy was associated with somewhat less responsiveness. These effects did not differ according to the conversation context, nor were they explained or moderated by relationship satisfaction. These findings suggest that empathic accuracy can facilitate responsive behavior, but only when perceivers are motivated to use their insight for benevolent goals. In the absence of empathic concern, empathic accuracy may be unhelpful (or even harmful).

In addition to this moderation effect, there was a strong main effect of empathic concern. At all levels of empathic accuracy, listeners who felt more emotional empathy were more responsive to their partners. This finding suggests that emotional empathy mobilizes caring behavior in close relationships (Canevello & Crocker, 2010; Collins et al., 2014) in much the same way that it fosters prosocial behavior among strangers (Batson, 1991). At the same time, our findings were clear in showing that empathic accuracy enabled compassionate listeners to be even more responsive than they would otherwise be. Overall, listeners were kinder and more responsive to their romantic partners when they felt greater emotional empathy during their conversations, but empathic accuracy further boosted their ability to effectively respond to their partner’s needs.

These findings have important theoretical implications for the study of responsiveness and social support. Consistent with responsiveness theory (Reis & Shaver, 1988), these findings provide the first evidence that accurate understanding and caring motivation are both necessary for optimal responsiveness. Even the most compassionate listeners were handicapped when they failed to accurately understand their partner’s thoughts and feelings. We speculate that empathic concern fosters expressions of caring, whereas empathic accuracy affords listeners the opportunity to express authentic understanding and validation (Reis & Gable, 2015). Our findings are also consistent with interpersonal models of social support, which suggest that the provision of sensitive and responsive support requires a combination of skills, resources, and benevolent motivation (e.g., Collins et al., 2010; Feeney & Collins, 2015). Our study points to empathic accuracy (a cognitive skill or resource) and empathic concern (benevolent motivation) as two key ingredients. In future research, it would be valuable to manipulate empathic accuracy and empathic concern to establish causal links. It would also be important to explore other aspects of responsive behavior, such as its paralinguistic or nonverbal components (e.g., turn taking, gaze, gesticulation; Duncan & Fiske, 1979).

The current study also has important implications for the literature on empathy. Our data showed that empathic accuracy and empathic concern were largely unrelated to each other (r = .15), which provides further evidence for the distinction between cognitive and affective forms of empathy and suggests that factors leading to accurate social perceptions may differ from those leading to compassionate feelings (Batson, 2009). The low correlation also suggests that these two forms of empathy do not have a simple causal effect on one another. Although it is often assumed that understanding another person’s thoughts and feelings will inspire compassion and prosocial motives, theoretical models of empathic accuracy take into account that accurate insight can sometimes foster egocentric motivation and behavior, especially in interdependent relationships in which mind reading can pose a threat to one’s self or one’s relationships (Ickes, in press; Simpson et al., 2003). Our data hint at this possibility. When empathic concern was low, higher empathic accuracy was associated with less responsiveness.

The impact of emotional empathy on empathic accuracy may be equally complex. Prior research shows that trait empathy does not increase empathic accuracy (e.g., Ickes, Stinson, Bissonnette, & Garcia, 1990). We extend this work by showing that in-the-moment feelings of empathy also fail to increase accuracy. One reason may be that although emotional empathy might inspire individuals to pay attention to their partners—to “tune in” to understand their partner’s inner states—perceivers may lack the skills or resources needed to do so. For example, perceivers may use flawed information, such as stereotypic knowledge of their partner’s past behavior, which can lead to incorrect inferences. Empathic accuracy may also be affected by features of the target, such as his or her ability to communicate clearly. Indeed, growing evidence suggests that empathic accuracy depends not only on features of the perceiver but also on the readability of the target (e.g., Ickes, Buysse, et al., 2000; Zaki, Bolger, & Ochsner, 2008). The current study (along with prior work) suggests no simple causal relationship between affective and cognitive forms of empathy. These findings point to exciting avenues for future research, thus highlighting the value of integrating research on empathy, prosocial behavior, and close relationships.

Conclusion

Understanding the thoughts and feelings of other people is an essential part of social interaction, but does this insight foster kindness and interpersonal sensitivity? Not necessarily. When it comes to responsive behavior, accurate understanding is clearly not sufficient on its own. Optimal responsiveness requires both accurate understanding and compassionate goals. Given the importance of responsiveness for relational well-being, we hope that our findings will inspire additional research on the complex links between cognitive and affective forms of empathy and their implications for prosocial behavior in close relationships.

Footnotes

Acknowledgements

We are deeply grateful to William Ickes for his invaluable conceptual and methodological guidance, to the many research assistants who collected and coded the data, and to the Nancy Collins/Shelly Gable Lab Group for helpful input on the study.

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Supplemental Material

Additional supporting information can be found at

Open Practices

All data and materials have been made publicly available via Open Science Framework and can be accessed at https://osf.io/fkmbh/. The complete Open Practices Disclosure for this article can be found at http://pss.sagepub.com/content/by/supplemental-data. This article has received badges for Open Data and Open Materials. More information about the Open Practices badges can be found at https://osf.io/tvyxz/wiki/1.%20View%20the%20Badges/ and .

Notes

References

Batson

C. D.

(1991). The altruism question: Toward a social-psychological answer. Hillsdale, NJ: Erlbaum.

Batson

C. D.

(2009). These things called empathy: Eight related but distinct phenomena. In Decety

Ickes

(Eds.), The social neuroscience of empathy (pp. 3–15). Cambridge, MA: MIT Press.

Batson

C. D.

Early

Salvarani

(1997). Perspective taking: Imagining how another feels versus imaging how you would feel. Personality and Social Psychology Bulletin, 23, 751–758.

Canevello

Crocker

(2010). Creating good relationships: Responsiveness, relationship quality, and interpersonal goals. Journal of Personality and Social Psychology, 99, 78–106.

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.

Collins

N. L.

Ford

Guichard

Kane

H. S.

Feeney

B. C.

(2010). Responding to need in intimate relationships: Social support and caregiving processes in couples. In Mikulincer

Shaver

P. R.

(Eds.), Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 367–388). Washington, DC: American Psychological Association.

Collins

N. L.

Kane

H. S.

Metz

M. A.

Cleveland

Khan

Winczewski

. . . Prok

(2014). Psychological, physiological, and behavioral responses to a partner in need: The role of compassionate love. Journal of Social and Personal Relationships, 31, 601–629.

Crocker

Canevello

(2008). Creating and undermining social support in communal relationships: The role of compassionate and self-image goals. Journal of Personality and Social Psychology, 95, 555–575.

Davis

M. H.

Kraus

L. A.

(1997). Personality and empathic accuracy. In Ickes

W. J.

(Ed.), Empathic accuracy (pp. 144–168). New York, NY: Guilford Press.

10.

Decety

Ickes

(Eds.). (2009). The social neuroscience of empathy. Cambridge, MA: MIT Press.

11.

Duncan

Fiske

D. W.

(1979). Dynamic patterning in conversation: Language, paralinguistic sounds, intonation, facial expressions, and gestures combine to form the detailed structure and strategy of face-to-face interactions. American Scientist, 67, 90–98.

12.

Feeney

B. C.

Collins

N. L.

(2003). Motivations for caregiving in adult intimate relationships: Influences on caregiving behavior and relationship functioning. Personality and Social Psychology Bulletin, 29, 950–968.

13.

Feeney

B. C.

Collins

N. L.

(2015). A new look at social support: A theoretical perspective on thriving through relationships. Personality and Social Psychology Review, 19, 113–147. doi:10.1177/1088868314544222

14.

Forest

A. L.

Kille

D. R.

Wood

J. V.

Holmes

J. G.

(2014). Discount and disengage: How chronic negative expressivity undermines partner responsiveness to negative disclosures. Journal of Personality and Social Psychology, 107, 1013–1032.

15.

Gable

S. L.

Gosnell

C. L.

Maisel

N. C.

Strachman

(2012). Safely testing the alarm: Close others’ responses to personal positive events. Journal of Personality and Social Psychology, 103, 963–982.

16.

Galinsky

A. D.

Moskowitz

G. B.

(2000). Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and in-group favoritism. Journal of Personality and Social Psychology, 78, 708–724.

17.

Gilin

Maddux

W. W.

Carpenter

Galinsky

A. D.

(2013). When to use your head and when to use your heart: The differential value of perspective-taking versus empathy in competitive interactions. Personality and Social Psychology Bulletin, 39, 3–16.

18.

Hendrick

S. S.

(1988). A generic measure of relationship satisfaction. Journal of Marriage and the Family, 50, 93–98.

19.

Ickes

(1993). Empathic accuracy. Journal of Personality, 61, 587–610.

20.

Ickes

(Ed.). (1997). Empathic accuracy. New York, NY: Guilford Press.

21.

Ickes

(in press). Empathic accuracy: Judging thoughts and feelings. In Hall

J. A.

Schmid Mast

West

(Eds.), The social psychology of perceiving others accurately. Cambridge, England: Cambridge University Press.

22.

Ickes

Buysse

Pham

Rivers

Erickson

J. R.

Hancock

. . . Gesn

P. R.

(2000). On the difficulty of distinguishing “good” and “poor” perceivers: A social relations analysis of empathic accuracy data. Personal Relationships, 7, 219–234.

23.

Ickes

Gesn

P. R.

Graham

(2000). Gender differences in empathic accuracy: Differential ability or differential motivation? Personal Relationships, 7, 95–109.

24.

Ickes

Robertson

Tooke

Teng

(1986). Naturalistic social cognition: Methodology, assessment, and validation. Journal of Personality and Social Psychology, 51, 66–82.

25.

Ickes

Stinson

Bissonnette

Garcia

(1990). Naturalistic social cognition: Empathic accuracy in mixed-sex dyads. Journal of Personality and Social Psychology, 59, 730–742.

26.

Kilpatrick

S. D.

Bissonnette

V. L.

Rusbult

C. E.

(2002). Empathic accuracy and accommodative behavior among newly married couples. Personal Relationships, 9, 369–393.

27.

Kline

R. B.

(2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press.

28.

Laurenceau

J.-P.

Barrett

L. F.

Rovine

M. J.

(2005). The interpersonal process model of intimacy in marriage: A daily-diary and multilevel modeling approach. Journal of Family Psychology, 19, 314–323.

29.

Maisel

N. C.

Gable

S. L.

(2009). The paradox of received social support: The importance of responsiveness. Psychological Science, 20, 928–932.

30.

Maisel

N. C.

Gable

S. L.

Strachman

(2008). Responsive behaviors in good times and in bad. Personal Relationships, 15, 317–338.

31.

Muthén

L. K.

Muthén

B. O.

(2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

32.

Reis

H. T.

(2006). The Perceived Responsiveness to Self scale. Unpublished manuscript, Department of Clinical and Social Sciences in Psychology, University of Rochester, Rochester, NY.

33.

Reis

H. T.

(2014). Responsiveness: Affective interdependence in close relationships. In Mikulincer

Shaver

P. R.

(Eds.), Mechanisms of social connection: From brain to group (pp. 255–271). Washington, DC: American Psychological Association.

34.

Reis

H. T.

Clark

M. S.

(2013). Responsiveness. In Simpson

J. A.

Campbell

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

35.

Reis

H. T.

Clark

M. S.

Holmes

J. G.

(2004). Perceived partner responsiveness as an organizing construct in the study of intimacy and closeness. In Mashek

D. J.

Aron

A. P.

(Eds.), Handbook of closeness and intimacy (pp. 201–225). Mahwah, NJ: Erlbaum.

36.

Reis

H. T.

Gable

S. L.

(2015). Responsiveness. Current Opinion in Psychology, 1, 67–71.

37.

Reis

H. T.

Patrick

B. C.

(1996). Attachment and intimacy: Component processes. In Kruglanski

Higgins

E. T.

(Eds.), Social psychology: Handbook of basic principles (pp. 523–563). New York, NY: Guilford Press.

38.

Reis

H. T.

Shaver

(1988). Intimacy as an interpersonal process. In Duck

(Ed.), Handbook of personal relationships (pp. 367–389). Chichester, England: Wiley.

39.

Sillars

A. L.

(1985). Interpersonal perception in relationships. In Ickes

(Ed.), Compatible and incompatible relationships (pp. 277–305). New York, NY: Springer-Verlag.

40.

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.

41.

Simpson

J. A.

Kim

J. S.

Fillo

Ickes

Rholes

W. S.

Oriña

M. M.

Winterheld

H. A.

(2011). Attachment and the management of empathic accuracy in relationship-threatening situations. Personality and Social Psychology Bulletin, 37, 242–254.

42.

Simpson

J. A.

Oriña

M. M.

Ickes

(2003). When accuracy hurts, and when it helps: A test of the empathic accuracy model in marital interactions. Journal of Personality and Social Psychology, 85, 881–893.

43.

Sprecher

Fehr

(2005). Compassionate love for close others and humanity. Journal of Social and Personal Relationships, 22, 629–651.

44.

Stinson

Ickes

(1992). Empathic accuracy in the interactions of male friends versus male strangers. Journal of Personality and Social Psychology, 62, 787–797.

45.

Tabachnick

B. G.

Fidell

L. S.

(2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education.

46.

Verhofstadt

L. L.

Buysse

Ickes

Davis

Devoldre

(2008). Support provision in marriage: The role of emotional similarity and empathic accuracy. Emotion, 8, 792–802.

47.

Zaki

Bolger

Ochsner

(2008). It takes two: The interpersonal nature of empathic accuracy. Psychological Science, 19, 399–404.

48.

Zaki

Weber

Bolger

Ochsner

(2009). The neural bases of empathic accuracy. Proceedings of the National Academy of Sciences, USA, 106, 11382–11387.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.60 MB

0.18 MB