Empathy Mediates the Link Between Extraversion and Latent Semantic Similarity in Computer-Mediated Discussions of a Politically Contentious Topic: Evidence from Abortion-Related Discourse

Abstract

Conversations about politically contentious issues often break down due to a lack of shared understanding, with language playing a central role in this failure—particularly in discussions of abortion, one of the most polarizing topics in U.S. public life. Mutual understanding requires that speakers achieve high latent semantic similarity (LSS) by using words in similar ways. While extraversion has been linked to higher LSS, likely due to its association with social engagement and attentional focus, semantic alignment also depends on speakers’ capacity for empathic attunement, suggesting that empathy may serve as a key mechanism linking extraversion to LSS. This study examined whether empathy mediates the relationship between extraversion and LSS in computer-mediated conversations between unacquainted dyads (N = 170) discussing the politically contentious topic of abortion. Results showed that empathy fully mediated this relationship, indicating that empathic attunement—not sociability alone—drives LSS. Findings inform communication theories by underscoring empathy's central role in fostering common ground understanding and highlight LSS as an observable linguistic marker of empathic engagement.

Keywords

latent semantic similarity latent semantic analysis dyadic interaction computer-mediated communication political discussions

Introduction

In the digital age, political discourse increasingly unfolds in online spaces, and social media have become a central platform for civic engagement. While these environments expand access to political dialogue, they also frequently give rise to interpersonal conflict—especially during discussions of politically contentious issues (Dunn, 2019; Willard, 2007). Such conflict often stems from a lack of mutual understanding between participants (Ensley et al., 2002; Hinds & Bailey, 2003; Koerner & Fitzpatrick, 2002; Leith & Baumeister, 1998), and language plays a pivotal role in this breakdown. Shared understanding depends on communicators using words in similar ways and with similar meanings (Bayram & Ta, 2020; Rossignac-Milon et al., 2021). Latent Semantic Similarity (LSS) captures this alignment by measuring the semantic overlap between individuals’ language use (Babcock et al., 2014; Ta et al., 2017).

Prior research identifies extraversion as a predictor of higher LSS, likely due to the trait's association with social engagement and attentional focus (Ta & Ickes, 2020). However, semantic alignment requires more than attentiveness and sociability—it hinges on a speaker's capacity for empathic attunement. That is, truly “getting on the same page” requires the ability to understand, resonate with, and adapt to a partner's thoughts, emotions, and communicative intentions. Extraverts may be more socially engaged, but without sufficient empathy, they may still fail to achieve meaningful alignment in conversation. This suggests that empathy—the capacity to understand and respond to others’ mental and emotional states (Cohen & Strayer, 1996; Van Langen et al., 2014)—may serve as a key mechanism linking extraversion to LSS. Yet, to date, no studies have directly tested this hypothesis.

The present study addresses this gap by investigating whether empathy mediates the relationship between extraversion and LSS in computer-mediated conversations between strangers discussing a highly contentious political topic: abortion. This question is especially important in light of the increasingly toxic nature of online discourse about politically contentious issues (Webster et al., 2022). Digital platforms have become fertile ground for incivility and harmful interpersonal dynamics (Rossini, 2022), with negative experiences resulting from discussions of politically contentious topics rising sharply in recent years (Cinelli et al., 2021; Su et al., 2018). Such discourse has been linked to heightened political polarization, increased intergroup hostility (Pew Research Center, 2019; Wojcieszak et al., 2015), and adverse health outcomes (Chavez et al., 2019). Thus, understanding the social psychological processes—such as empathy—that facilitate mutual understanding in conversations focused on a politically contentious subject is essential.

In the present study, we focus on extraversion not as a direct or exclusive predictor of semantic alignment, but as a distal antecedent of empathy. Extraversion reflects a social approach orientation characterized by interpersonal engagement, affiliative motivation, and heightened sensitivity to social reward (Wilt & Revelle, 2009). Although other socially relevant traits—such as agreeableness or openness—may support constructive interaction, extraversion is particularly tied to the motivation to initiate, sustain, and attend to interpersonal exchanges during first-time interactions. These socio-motivational tendencies are theoretically relevant because they increase individuals’ likelihood of engaging empathically with a conversational partner, which in turn may support the development of shared meaning in dialogue. Accordingly, we conceptualize extraversion as an upstream personality variable that contributes to LSS indirectly through empathy, rather than as a strong stand-alone predictor of semantic alignment. This framing distinguishes the present study from prior work by shifting the focus from identifying trait correlates of LSS to testing a socio-cognitive mechanism that explains how extraversion relates to shared meaning.

Background Literature

Interaction partners rely on language to develop a “common-ground understanding” (e.g., Abbeduto et al., 1998; Kecskes & Zhang, 2009) or an “intersubjective meaning context” (e.g., Gesn & Ickes, 1999; Morganti, 2008) with one another. This enables interaction partners to coordinate their behaviors and more effectively navigate through the interaction to establish shared meaning (Higgins et al., 2021; Rossignac-Milon et al., 2021). LSS has been used to empirically measure this alignment across many domains, including international relations negotiations (Bayram & Ta, 2020), patient-physician communication (Vrana et al., 2018), shared reality (Rossignac-Milon et al., 2021), and others. LSS is derived through latent semantic analysis (LSA), a computational technique that examines word co-occurrence patterns to infer semantic relationships within a body of text (Landauer et al., 1998; Landauer & Dumais, 1997). To calculate LSS, LSA is used to assess the semantic content of two separate text segments by mapping them into a multidimensional semantic space. The similarity between the two texts is then determined by computing the cosine of the angle between their corresponding vectors, yielding a score between −1 and 1, with higher values reflecting greater semantic alignment. Rather than measuring whether the same words are used, LSS evaluates how similarly words are used in context, which captures deeper, conceptual similarities in meaning. This makes LSS a more nuanced indicator of mutual understanding than other linguistic metrics such as those focusing on lexical similarity (Babcock et al., 2014).

Personality traits have been consistently linked to the quality and dynamics of social interactions (e.g., Asendorpf & Wilpers, 1998; Back et al., 2010). Extraversion is particularly influential in facilitating successful exchanges (e.g., Mehl et al., 2006; Wilt & Revelle, 2009) given that it is marked by sociability, talkativeness, and an outwardly focused orientation (Jensen, 2016). Individuals high in extraversion tend to be more engaged with their interaction partners and more attentive to social cues (Blevins et al., 2022), suggesting that this trait may contribute to stronger interpersonal alignment as reflected in LSS. This has been empirically supported in several studies. Babcock et al. (2014) found that LSS was more likely to develop between partners who were mutually attentive and highly engaged in conversation—particularly when rich verbal content was exchanged. Similarly, Ta et al. (2017) demonstrated that behaviors such as speaking, maintaining eye contact, and acknowledging one's partner facilitated the development of dyadic LSS. Ta and Ickes (2020) found that in initial dyadic computer-mediated conversations, participants were most motivated to achieve high levels of LSS during the early stages of the exchange. For most dyads, this motivation declined over time and resulted in a gradual decrease in LSS. However, this pattern did not hold for dyads composed of highly extraverted individuals, whose LSS levels remained consistently high throughout the conversation. Elevated extraversion appeared to sustain partners’ mutual focus and engagement across the full interaction—not only due to extraverts’ outward attentional orientation, but also because of their heightened need for external stimulation, which a new conversational partner can provide. Together, these findings not only suggest that more verbally expressive individuals play a greater role in fostering and maintaining shared meaning, but they also highlight the role of motivational and attentional processes in shaping LSS. This suggests that other traits—particularly those that support the attentional and motivational processes underlying LSS, such as empathy—may also play a key role in fostering shared understanding during conversation.

Empathy is widely regarded as a cornerstone of social understanding, enabling individuals to accurately perceive and interpret their interaction partner's thoughts and emotions (Ickes, 2010). Beyond fostering emotional resonance, empathy plays a critical role in shaping communicative behavior by allowing individuals to adapt their language and responses in ways that enhance alignment, coordination, and the construction of common ground. This perspective is reinforced by a range of influential communication theories. Frameworks such as the encoding–decoding model (Hall, 1993), the intentionalist model (Motley, 1986), the perspective-taking paradigm (Krauss, 2002), and the dialogic model (Clark & Brennan, 1991) converge on a shared premise: that communication is a joint, interpretive process that is optimized when individuals are both motivated and capable of empathizing with their partner's cognitive and emotional states. Across these models, empathy emerges not merely as a moral virtue or affective trait, but as a central cognitive-affective mechanism that is essential for decoding messages, inferring communicative intent, adopting others’ perspectives, and co-constructing shared meaning.

Affective empathy (rather than cognitive empathy) is especially important because it is more directly linked to emotional attunement, interpersonal warmth, and prosocial motivation (Haas et al., 2015; Van Langen et al., 2014)—traits that are especially relevant in emotionally charged or morally polarized contexts such as abortion-related discussions. Prior research also suggests that affective empathy plays a critical role in fostering interpersonal connection and reducing perceived threat during disagreement (e.g., Decety & Cowell, 2014; Zaki, 2014). Moreover, affective empathy is often more spontaneous and less cognitively demanding than its cognitive counterpart, making it particularly salient in real-time, naturalistic interactions where emotional dynamics can unfold quickly.

Empirical work shows that individuals higher in extraversion often report greater empathic concern and engage more readily in behaviors associated with interpersonal warmth and attunement (Airagnes et al., 2021; Bertram et al., 2016; Jolliffe & Farrington, 2006). These associations do not imply that empathy is a stable personality trait in the same sense as extraversion; rather, they suggest that extraverted individuals may enter social interactions with a stronger tendency or motivation to engage emotionally with others. In interaction, empathy is expressed situationally through moment-to-moment processes, such as attending to a partner's cues, interpreting emotional signals, and adapting communicative behavior, that support shared meaning. Thus, while extraversion functions as a broad dispositional orientation toward social engagement, empathy reflects the real-time interpersonal attunement through which semantic alignment can emerge during conversation. This distinction is central to our model: extraversion may set the stage for engagement, but empathy is the mechanism through which individuals align their language with a partner.

Although other socially relevant traits, such as warmth, gregariousness, assertiveness, excitement-seeking, and positive emotion, could provide insights into how LSS forms, extraversion is particularly important because it integrates these tendencies into a broader social-approach orientation (Lucas et al., 2000; Watson et al., 2015). Accordingly, extraversion is widely recognized as a primary driver of social behavior (Eaton & Funder, 2003; Fleeson & Gallagher, 2009; Sherman et al., 2015), orienting individuals toward outward engagement and interpersonal exchange (Breil et al., 2019). In this view, extraversion contributes to social engagement and attentional focus while empathy enables the deeper interpersonal attunement necessary for the development of shared meaning in conversation. As such, empathy would mediate the relationship between extraversion and LSS by translating social motivation into communicative alignment. Further, extraversion, empathy, and willingness to communicate are uniquely related (Zohoorian et al., 2022), providing an important theoretical background for the use of talking and LSS in this study.

LSS offers a linguistic indicator of the extent to which conversational partners converge on a shared meaning framework during interaction. Although LSS does not capture the full scope of shared reality, it reflects a fundamental component of common ground understanding: the alignment of conceptual representations that make language interpretable and mutually meaningful (Echterhoff et al., 2009; Higgins et al., 2021). Shared meaning is especially consequential in discussions of politically contentious issues, which often involve competing moral framings and divergent assumptions about facts, values, and identities. When interaction partners fail to establish a minimum level of semantic common ground, conversations may stall, become adversarial, or produce talking-past-one-another dynamics. Because LSS quantifies the degree to which partners use language in increasingly similar ways over the course of a dialogue, it offers an ecologically relevant lens for examining the development of mutual understanding in discussions of politically contentious issues where misunderstandings and divergent interpretive frames often impede constructive communication.

The Current Study

In this study, we examined the mediating role of empathy on the relationship between extraversion and LSS during dyadic computer-mediated conversations between unacquainted partners discussing a highly controversial political topic—abortion. Specifically, we examined whether extraversion predicts higher LSS indirectly through increased empathy (Figure 1). This mediation model posits that extraversion positively influences empathy (path a), which, in turn, positively predicts LSS (path b), and that the total effect of extraversion on LSS (path c) would be explained by this indirect pathway. We also estimated the direct effect of extraversion on LSS when controlling for empathy (path c′) to assess whether the relationship is fully or partially mediated. By testing this model, we aim to clarify the psychological mechanisms through which individual differences in personality contribute to shared meaning and common ground understanding in online discussions about a politically contentious topic: abortion.

Figure 1.

Conceptual mediation model: empathy as a mediator of the relationship between extraversion and LSS. Note. This figure demonstrates the hypothesized indirect effect of extraversion (X) on LSS (Y) through empathy (M). Path a represents the effect of extraversion on empathy. Path b represents the effect of empathy on LSS controlling for extraversion. Paths c and c′ represent the total and direct effects of extraversion on LSS, respectively. Messaging frequency and gender were included as covariates throughout the entire model.

Abortion was selected as the discussion topic because it represents a paradigmatic case of morally charged and identity-relevant political conflict in the United States. Unlike many policy disagreements, the topic of abortion often evokes strong moral convictions, deep affective responses, and stable attitudinal positions that are closely tied to social identity and group affiliation (Pew Research Center, 2022). These features make the topic of abortion an appropriate test case for examining the psychological processes that enable or inhibit the formation of shared meaning in dialogue about a politically contentious topic. If LSS depends on empathic attunement, then such effects should be particularly evident when interaction partners are navigating an issue where mutual understanding is often fragile and communicative breakdown is common. Thus, the use of abortion provides a theoretically rich and ecologically valid context in which to investigate linguistic mechanisms relevant to understanding in discussions on a politically contentious topic.

Initial dyadic interactions offer a unique and powerful lens through which to examine the psychological processes that underlie the formation of common ground, especially when partners have no prior relationship history (Ickes, 2010). In such first-time encounters, the absence of shared background knowledge or established relational dynamics allows researchers to more clearly isolate the effects of dispositional traits (such as empathy and extraversion) on communicative outcomes (McDaniel & Coyne, 2016). Moreover, studying language in initial, computer-mediated conversations closely mirrors how many conversations about a contentious political topic, including abortion, unfold in contemporary society: through text-based interactions between strangers in online forums and social media platforms (Bail, 2022; Vaccari & Valeriani, 2021). In addition, text-based computer-mediated environments, compared to face-to-face exchanges, attenuate nonverbal cues and heighten ambiguity, which places greater demands on linguistic strategies to establish mutual understanding (Baek et al., 2012). These features make such contexts especially well-suited for examining LSS as the entire interaction is verbal and the language used tends to be more intentional and cognitively filtered than in spontaneous, face-to-face conversation (Placiński & Żywiczyński, 2023; Sacristan et al., 2023).

Although this study is situated within research on polarized political communication, it does not directly measure polarization or changes in political attitudes. Rather, the present work focuses on a theoretically relevant communicative mechanism, LSS, within online conversations about an issue that is deeply intertwined with affective and ideological polarization in the United States. Our goal is not to claim that LSS reduces polarization, but to identify a linguistic process that may facilitate mutual understanding in dialogue about a politically contentious topic. To systematically examine this, interaction partners were randomly assigned to include either dyads who agreed or disagreed on the issue. This mix of attitudinal congruence and incongruence enables a more nuanced understanding of how empathy facilitates mutual understanding across variations of ideological common ground. By combining a socially relevant topic, an ecologically valid communication medium, and a structured dyadic design, this approach offers a rigorous test of the psychological and linguistic mechanisms that enable—or inhibit—the development of common ground understanding in online discussions on abortion.

Method

Participants

All participants (N = 340) were at least 18 years of age and were undergraduate students from a large university in the southwest United States. The sample consisted of 70.59% women and 29.41% men. Ages ranged from 18 to 56 (M = 21.04, SD = 4.69). Regarding race/ethnicity, 30.82% identified as White, 25.29% identified as Hispanic/Latino, 21.47% identified as Asian, 17.35% identified as Black, 4.12% identified as Other/Multiracial, 0.59% identified as Native Hawaiian or Pacific Islander, and 0.29% identified as Native American or Alaskan Native.

Procedure

Participants were recruited through a departmental subject pool. During the pre-screening process, participants responded to two questions assessing their views on abortion. This was then used to categorize participants as either pro-life or pro-choice (see Materials section). Then, each participant was randomly paired with another participant of the same sex who either shared or did not share their stance on abortion (hereafter referred to as agreeing dyads and disagreeing dyads, respectively). Participants were not informed of their pairing type. The final sample consisted of 94 agreeing dyads and 76 disagreeing dyads for a total of 170 dyads. The Supplementary Materials report additional descriptive characteristics of the sample across agreement and abortion stance.

Dyads were then scheduled to participate in the study at a designated date and time. Each dyad member was instructed to arrive at different locations to ensure that they did not see or interact with each other before the study. After providing informed consent, they completed a pre-interaction survey in which they provided demographic information and completed the Big Five Inventory-10 (BFI-10; Rammstedt & John, 2007) and the Toronto Empathy Questionnaire (TEQ; Spreng et al., 2009). Participants also indicated their stance on abortion to ensure that they endorsed the same position on abortion as they did during the pre-screening process.

After both dyad members completed the survey, they were instructed to discuss their respective views regarding the topic of abortion with another participant in the study via instant messenger for 18 min. Dyad members did not have prior knowledge of their interaction partner's stance on abortion. Dyad members exchanged an average of M = 14.76 messages (SD = 8.42), resulting in conversations that averaged M = 254.14 words (SD = 79.57). Upon completion of their discussion, each dyad member completed a post-interaction survey in which they indicated their own stance on abortion and inferred their dyad partner's stance on abortion. This was done to determine if participants endorsed the same stance as they did before their discussion and if they accurately inferred their dyad partner's stance on abortion. All dyad members were then debriefed, confirmed that they did not know the identity of their interaction partner, and granted permission for their discussion transcript to be saved and used for data analysis. Participants answered several additional questions in the pre- and post-interaction surveys that were not pertinent to the current investigation and were thus excluded from analysis.

The transcripts were then used to calculate each dyad's LSS index using the same method in previous studies (e.g., Babcock et al., 2014; Ta et al., 2017; Ta & Ickes, 2020). For a given transcript, each dyad member's speaking turn was separated into two blocks of text, each of which contained only one dyad member's portion of their discussion. The two blocks of text were then analyzed using the LSA Pairwise Comparison program (Landauer et al., 1998), which generated a single LSS score for the given dyad. This method computes the cosine of the angles between the two resulting vectors to estimate LSS based on the words that are used in the transcript and how those words are used in relation to other words in the transcript (Landauer et al., 1998). This process was repeated for each transcript, resulting in an LSS score for each dyad. Higher LSS scores indicated greater semantic similarity. All study procedures were approved by the Institutional Review Board.

Materials

Pre-Screening

Abortion Stance. Participants rated the degree to which they agreed with the following statements on a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree): Abortion should be legal and local healthcare providers should provide legal abortions. Both items demonstrated excellent internal consistency (inter-item correlation, r = .97, p < .001). Participants who agreed and/or strongly agreed with both items were classified as pro-choice, whereas participants who disagreed and/or strongly disagreed with both items were classified as pro-life.¹ This was used to form agreeing dyads (i.e., dyads in which both partners were pro-choice or pro-life) and disagreeing dyads (i.e., dyads in which one partner was pro-life and the other was pro-choice). Individuals who were neutral on one or both items or agreed/strongly agreed on one item but disagreed/strongly disagreed on the other item during the pre-screening process were not included in the study. This method was adapted from previous psychological studies that classified participants as pro-life or pro-choice (e.g., Chambers et al., 2006; Hartmann & Tanis, 2013).

Pre-Interaction Survey

Demographics. Participants provided demographic information, including age, gender, and race/ethnicity.

Abortion Stance. Participants rated the degree to which they agreed with the following statements on a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree): Abortion should be legal and local healthcare providers should provide legal abortions. Both items demonstrated excellent internal consistency in the pre-screening survey (inter-item correlation, r = .97, p < .001) and pre-interaction survey (inter-item correlation, r = .98, p < .001). The items also demonstrated excellent internal consistency with one another across the pre-screening and pre-interaction surveys (inter-item correlations, r = .94–.95, ps < .001), indicating that participants endorsed a consistent stance on abortion across both time points (more information can be found in the Supplementary Materials).

BFI-10. The BFI-10 (Rammstedt & John, 2007) is a short 10-item scale that measures the Big Five dimensions of personality: Conscientiousness, Agreeableness, Neuroticism, Openness, and Extraversion. Items are scored on a 5-point fully anchored strongly disagree/strongly agree response set. Only the items measuring extraversion were used in the current study. These items demonstrated acceptable internal consistency in the current sample (α = .71).

TEQ. The TEQ (Spreng et al., 2009) is a 16-item scale that measures the emotional (i.e., affective) components of empathy. The items represent a wide variety of affective empathy-related behaviors such as emotion comprehension, sympathetic physiological arousal, and altruism. Items are scored on a 5-point fully anchored never/always response set with higher scores indicating a greater degree of empathy. The scale demonstrated good internal consistency (α = .80) in the current sample.

Post-Interaction Survey

Self-Reported and Perceived Partner Abortion Stance. Participants rated the extent to which they personally believed abortion should be legal (self-reported abortion stance) and the extent to which they believed their dyad partner supported the legality of abortion (perceived partner abortion stance). Both items were rated on a 7-point Likert-type scale (1 = abortion should not be legal, 7 = abortion should be legal). Each participant's self-reported abortion stance and their dyad partner's perception of their abortion stance were strongly correlated, r(338) = .82, p < .001, suggesting that participants were able to accurately infer their dyad partner's abortion stance. Self-reported abortion stance demonstrated excellent internal consistency with the abortion stance items assessed during pre-screening (inter-item correlations, rs = .88–.97, ps < .001), and the pre-interaction survey screening (inter-item correlations, rs = .89–.98, ps < .001), indicating that participants endorsed a consistent stance on abortion across all 3 time points. This was also observed among agreeing dyads and disagreeing dyads (see Supplementary Materials for more information).

Results

Composite scores for empathy and extraversion were computed for each participant and then averaged across dyad members to yield dyad-level empathy scores and dyad-level extraversion scores (hereafter referred to as simply empathy and extraversion) as consistent with prior research (Ta et al., 2017; Ta & Ickes, 2020). Because LSS is calculated at the dyad level, this aggregation aligns with the level of analysis. The number of messages exchanged has been shown to predict LSS (Ta & Ickes, 2020). Accordingly, the average number of messages exchanged within each dyad (M = 14.76, SD = 8.42) was computed to generate a single messaging frequency score for each dyad.

Descriptive statistics and zero-order correlations of key variables are reported in Table 1. Extraversion and empathy were positively associated with LSS, and extraversion was positively associated with empathy. A multiple regression model was first conducted using the lm function in RStudio (R Core Team, 2024; Version 4.3.3) with LSS as the outcome and both extraversion and empathy entered as predictors. Dyad gender composition (male-male dyads, female-female dyads), agreement (agreeing dyads, disagreeing dyads), and messaging frequency were considered as potential control variables. Welch's t tests indicated that LSS did not significantly differ across gender composition, t(168) = 2.81, p = .10, or agreement, t(168) = 1.41, p = .24. A Pearson correlation indicated that messaging frequency was significantly correlated with LSS, r(168) = .27, p < .001. As such, only messaging frequency was entered into the multiple regression model as a control variable to maintain model parsimony and minimize overfitting. Multicollinearity was assessed using tolerance values and variance inflation factors (VIF). All variables exhibited acceptable collinearity statistics (all VIF values <2), suggesting no significant multicollinearity issues. Additional descriptive characteristics for the sample can be found in the Supplementary Materials.

Table 1.

Descriptive Statistics and Zero-Order Correlations.

Variable	M (SD)	LSS	Extraversion	Empathy	Messaging frequency
LSS	0.73(0.08)	–
Extraversion	3.09(0.72)	.18*	–
Empathy	4.12(0.28)	.24**	.24**	–
Messaging frequency	14.76(8.42)	.27***	.20**	.03	–

Note. All df = 168; M = mean; SD = standard deviation.

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

The overall multiple regression model significantly predicted LSS, F(3, 166) = 8.17, p < .001, adjusted R² = .11 (Table 2). Empathy significantly predicted LSS such that higher levels of empathy predicted higher LSS, β = .21, t = 2.84, SE = 0.02, p = .005. However, extraversion did not significantly predict LSS, β = .08, t = 1.04, SE = 0.01, p = .30. This pattern suggests that empathy accounts for the variance in LSS previously attributed to extraversion and provides justification to conduct a mediation analysis.

Table 2.

Results of Multiple Regression Model Predicting LSS.

Predictor	β (SE)	t
Extraversion	.01(0.01)	1.04
Empathy	.02(0.01)	2.84**
Messaging frequency	.02(0.01)	3.27**

Note. β = standardized regression coefficient; SE = standard error; t = t statistic.

**p < .01.

A mediation analysis was conducted using the rosetta package in RStudio (Peters & Verboon, 2023; Version 0.3.12) to test whether empathy mediated the relationship between extraversion and LSS controlling for messaging frequency and dyad gender composition given that women tend to score higher on empathy than men (e.g., Pang et al., 2023; Wu et al., 2023) (Figure 1). The direct effect indicates the degree to which the predictor (extraversion) is linked with the outcome (LSS) controlling for the mediator (empathy), and the total effect indicates the degree to which the predictor is linked with the outcome not controlling for the mediator. The indirect effect indicates the degree to which the predictor is linked with the potential mediating variable as well as the degree to which the mediator is linked with the outcome. Mediation is present when the indirect effect is significant.

Standardized path coefficients (β), which function as completely standardized effect sizes, provide a meaningful index of pathway strength. Results indicated that extraversion was significantly associated with empathy (path a), β = .22, SE = 0.03, p = .002, and that empathy was significantly associated with LSS when controlling for extraversion (path b), β = .21, SE = 0.02, p = .007. The indirect effect of extraversion on LSS via empathy was significant, β = .05, SE = 0.002, p = .04, supporting a significant mediation effect. The direct effect of extraversion on LSS controlling for empathy was not significant (path c′), β = .07, SE = 0.01, p = .29, and the total effect was also not significant (path c), β = .13, SE = 0.01, p = .09, indicating that empathy fully mediated the relationship between extraversion and LSS.

Discussion

Extraversion has been associated with higher LSS likely due to its links with greater social engagement and attentional focus (Ta & Ickes, 2020). However, achieving semantic alignment in conversation also involves the capacity to understand, resonate with, and adapt to a partner's thoughts and emotions. This suggests that empathy may serve as a key mechanism through which extraversion promotes shared meaning in discussions. Despite this theoretical link, prior research has not directly tested empathy's mediating role. The current study addressed this gap by examining whether empathy mediates the relationship between extraversion and LSS in dyadic, computer-mediated conversations between unacquainted partners discussing a politically contentious issue: abortion. These findings provide new insight into the psychological mechanisms that support mutual understanding in discussions about a politically contentious issue—a topic of growing importance amid the sharp rise in negative experiences discussing politically contentious topics within digital environments in recent years (Cinelli et al., 2021; Rossini, 2022; Su et al., 2018; Webster et al., 2022). Although LSS reflects only one dimension of shared meaning, its emergence suggests that interaction partners are beginning to construct an overlapping interpretive context which may represent a necessary, though not sufficient, condition for productive engagement in conversation about a politically contentious issue.

The mediation analysis revealed that the relationship between extraversion and LSS was statistically explained by empathy: when empathy was included in the model, the direct effect of extraversion on LSS became non-significant. Although extraversion has been identified in prior work as a socially relevant antecedent, the total effect in the present study was small in magnitude, which aligns with earlier findings showing that extraversion does not consistently predict LSS directly (Babcock et al., 2014; Ta et al., 2017; Ta & Ickes, 2020). Thus, the present results should not be interpreted as evidence that extraversion is a strong or unique driver of semantic alignment. Instead, the contribution of this study lies in demonstrating that empathy operates as the more proximal socio-cognitive mechanism through which individuals achieve greater shared meaning. In other words, greater LSS is not primarily driven by sociability, expressiveness, or interpersonal engagement alone—as captured by extraversion—but by the ability to accurately perceive and respond to a partner's thoughts and emotions. This distinction clarifies the underlying mechanism linking extraversion to LSS and identifies empathy as the more psychologically consequential driver of shared meaning.

By foregrounding empathy rather than extraversion, the present study positions semantic alignment as an interpersonal attunement process rather than a personality-driven one. This perspective aligns with both the small empirical effect of extraversion in the current data and broader evidence that personality traits shape communication primarily through socio-cognitive mechanisms. While extraversion may increase individuals’ general motivation to engage with others, empathy appears critical for the moment-to-moment attunement and adaptation that enable partners to converge on a shared semantic space.

This distinction also clarifies the psychological processes emphasized by major theories of communication, which argue that successful interaction requires continuous coordination of meaning and sensitivity to a partner's intentions and understanding. The present findings indicate that this coordination is not simply a function of being sociable or talkative, but instead depends on core components of empathy, such as emotional attunement and perspective-taking. Empathic individuals are better equipped to monitor, interpret, and adapt to their partner's communicative needs, thereby fostering greater semantic alignment. This highlights empathy as a foundational social–cognitive capacity in the co-construction of meaning and positions LSS as a behavioral trace of empathic engagement in conversation. This interpretation is consistent with prior research showing that individuals who are more interpersonally attuned tend to exhibit greater linguistic alignment (Gonzales et al., 2010; Ireland & Pennebaker, 2010). Theoretical models of common-ground formation should therefore consider empathy as a central mechanism through which shared meaning emerges.

These findings also contribute to a small but growing body of research showing that individual differences in empathy play a central role in promoting civility and mutual understanding in discourse on politically contentious topics independent of whether individuals agree ideologically (Bejan, 2017; Saveski et al., 2022). While other studies primarily focus on the role of shared beliefs in fostering smoother interactions (e.g., Bright, 2016; Skytte, 2021; Van Elsas & Fiselier, 2024), our results indicate that empathy can facilitate respectful and meaningful dialogue even in the absence of attitudinal agreement. This suggests that common ground understanding is not solely a product of attitudinal alignment, but also of interpersonal motivation and skill. Empathic individuals may be more willing and able to listen carefully, consider opposing viewpoints, and communicate in ways that reduce defensiveness and conflict. In this way, empathy emerges as a dispositional resource that enables constructive engagement across attitudinal divides, offering a psychological explanation for why some conversations remain civil and productive despite deep disagreement. In addition, although higher LSS reflects greater common ground understanding, this does not imply that the conversation was enjoyable or subjectively positive. Semantic alignment captures how similarly partners are using language, not how they felt about the exchange; individuals can achieve high LSS even during conversations that are strained, effortful, or emotionally difficult.

Moreover, these findings carry several important practical implications for enhancing LSS and the quality of discourse on a politically contentious subject in both interpersonal and digital contexts. First, it identifies empathy, not merely extraversion, as a key driver of shared understanding, suggesting that interventions aimed at cultivating empathy may be more effective than those focused on promoting general sociability. Given that empathy is a malleable and trainable trait (Schumann et al., 2014; Zaki, 2014), it represents a promising target for a range of interventions designed to improve discourse focused on a politically contentious topic. For example, civic education programs can incorporate empathy-building modules that emphasize active listening, respectful disagreement, recognition of diverse lived experiences, and other skills essential for democratic engagement. Additionally, emotional attunement training, which is often used in clinical or conflict resolution settings, can help individuals recognize and respond appropriately to others’ emotional cues, fostering relational warmth even amid disagreement. These approaches offer scalable pathways for cultivating psychological capacities that facilitate higher LSS and more constructive discussions about a politically contentious topic. Second, the computer-mediated nature of this study highlights its relevance for online conversations about a politically contentious topic in which the absence of nonverbal cues and the prevalence of misinterpretation make empathic engagement particularly important. Digital platforms could integrate design features that encourage perspective-taking, such as reflective prompts or mechanisms that emphasize common ground, to foster greater LSS and more constructive and civil exchanges.

Third, the results suggest that empathy is not solely an internal emotional or dispositional state but is also externally expressed and measurable through linguistic behavior—in this case, through LSS. Rather than existing only as a private feeling or cognitive orientation, empathy appears to manifest in the ways individuals use language to align with others: adapting word choice, meaning structures, and conversational framing to facilitate mutual understanding. This reinforces the idea echoed in previous studies (Pennebaker & Graybeal, 2001; Zaki, 2014) that shared understanding is not merely a cognitive inference or subjective impression, but a dynamic process that is enacted in and through language. As such, language thus becomes a behavioral trace of empathy, offering observable evidence of interpersonal attunement and providing a language-based approach to studying empathy in naturalistic contexts that complements traditional self-report or physiological measures while advancing efforts to quantify empathic processes in real-world dialogue on a politically contentious topic.

Strengths, Limitations, and Future Directions

This study has several strengths. To our knowledge, it is the first to demonstrate that empathy mediates the association between extraversion and LSS in initial dyadic, computer-mediated conversations about a politically contentious issue. Importantly, we do not present extraversion as a strong or exclusive predictor of LSS; instead, we treat it as a theoretically meaningful antecedent for testing whether empathy functions as the proximal mechanism shaping semantic alignment. The use of computer-mediated conversation enhances ecological validity by mirroring the text-based environments in which discussions of abortion (and many other contentious public issues) now commonly occur. These contexts involve exchanges between individuals who may not share a relationship history or communicative norms, and they rely heavily on language in the absence of nonverbal cues. As such, they are well suited for examining linguistic alignment and common-ground formation using LSS.

By including dyads who either agreed or disagreed on abortion, the study also assesses whether empathy supports shared meaning across different levels of attitudinal agreement. Although this design does not measure changes in polarization, it provides an initial indication of whether interlocutors can achieve semantic alignment even when they hold opposing views. This approach enables a more nuanced examination of empathy as a relational process that may facilitate common-ground understanding independent of ideological similarity.

Using abortion as the conversation topic further strengthens the design. Abortion is one of the most morally and politically charged issues in the United States and reliably elicits strong emotional and ideological responses (Pew Research Center, 2022). These features place heightened demands on psychological capacities such as empathy and perspective-taking, offering a rigorous test of whether these traits are associated with semantic alignment during ideologically challenging dialogue. Because abortion is frequently debated in both online and offline settings, the findings are relevant for understanding how shared meaning may emerge or break down in real-world conversations about politically contentious issues, even though broader claims about polarization cannot be drawn from this design alone.

Several limitations should be noted. Focusing on a single, highly distinctive issue necessarily constrains generalizability. Abortion's moral intensity, its ties to religious and gender identities, and its centrality to U.S. political polarization may produce interpersonal dynamics different from those observed in conversations about issues such as climate change, immigration, or gun policy. Future work examining a broader range of politically contentious topics would help determine whether the empathy-LSS pathway generalizes across issues or varies with domain-specific features. In addition, although the sample was adequate for detecting medium-sized effects, larger and more diverse samples would provide greater power to test moderation and capture variation in age, ideology, and communicative norms.

Future research could also test whether empathy can be experimentally increased to enhance LSS. Brief interventions, such as perspective-taking prompts, empathy priming, or exposure to personal narratives, may heighten participants’ capacity to attend to and align with a partner's language. Longer-term programs, including structured empathy training or deliberative dialogue initiatives, could assess whether sustained improvements in empathic skill produce more durable gains in LSS and conversational quality. Such work would help clarify the causal role of empathy in fostering shared meaning and offer practical strategies for supporting more constructive engagement in digital environments, where empathic processing is often reduced.

Finally, longitudinal or repeated-interaction designs could illuminate how empathy and LSS develop, fluctuate, or stabilize over time. Unlike single-session conversations, ongoing interactions allow researchers to observe whether empathic attunement and semantic alignment strengthen through repeated exposure, mutual adaptation, or growing familiarity. These designs would also allow for a time lag between personality assessment and conversation, reducing potential priming effects. Repeated interactions would further enable researchers to examine whether empathy and LSS buffer against conversational breakdowns during moments of disagreement or emotional intensity. Such approaches would deepen understanding of the temporal dynamics of LSS and provide a more ecologically valid model of how conversations about politically contentious issues unfold in everyday life.

Supplemental Material

sj-docx-1-jls-10.1177_0261927X251414971 - Supplemental material for Empathy Mediates the Link Between Extraversion and Latent Semantic Similarity in Computer-Mediated Discussions of a Politically Contentious Topic: Evidence from Abortion-Related Discourse

Supplemental material, sj-docx-1-jls-10.1177_0261927X251414971 for Empathy Mediates the Link Between Extraversion and Latent Semantic Similarity in Computer-Mediated Discussions of a Politically Contentious Topic: Evidence from Abortion-Related Discourse by Vivian P. Ta-Johnson and Isabella M. Swafford in Journal of Language and Social Psychology

Footnotes

Ethics Approval

This study was approved by the Institutional Review Board at the University of Texas at Arlington (Approval No. 2017-0367.1) on January 12, 2017. Respondents gave written consent before participating in the study.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability

Data used in this study are available and can be accessed in the link below. The data set should be cited as follows (Ta-Johnson & Swafford, 2026).

Supplemental Material

Supplemental material for this article is available online.

ORCID iDs

Vivian P. Ta-Johnson

Isabella M. Swafford

Notes

Author Biographies

Vivian P. Ta-Johnson is an assistant professor of Psychology. Her area of research uses natural language processing to understand social psychological processes.

Isabella M. Swafford is a PhD student in Psychology. Her area of research uses natural language processing to examine linguistic markers of persuasion and mental health.

References

Abbeduto

Short-Meyerson

Benson

Dolish

Weissman

(1998). Understanding referential expressions in context: Use of common ground by children and adolescents with mental retardation. Journal of Speech, Language, and Hearing Research, 41(6), 1348–1362. https://doi.org/10.1044/jslhr.4106.1348

Airagnes

Du Vaure

C. B.

Galam

Bunge

Hoertel

Limosin

Jaury

Lemogne

(2021). Personality traits are associated with cognitive empathy in medical students but not with its evolution and interventions to improve it. Journal of Psychosomatic Research, 144, 110410. https://doi.org/10.1016/j.jpsychores.2021.110410

Asendorpf

J. B.

Wilpers

(1998). Personality effects on social relationships. Journal of Personality and Social Psychology, 74(6), 1531–1544. https://doi.org/10.1037/0022-3514.74.6.1531

Babcock

M. J.

V. P.

Ickes

(2014). Latent semantic similarity and language style matching in initial dyadic interactions. Journal of Language and Social Psychology, 33(1), 78–88. https://doi.org/10.1177/0261927X13499331

Back

M. D.

Schmukle

S. C.

Egloff

(2010). Why are narcissists so charming at first sight? Decoding the narcissism–popularity link at zero acquaintance. Journal of Personality and Social Psychology, 98(1), 132–145. https://doi.org/10.1037/a0016338

Baek

Y. M.

Wojcieszak

Delli Carpini

M. X.

(2012). Online versus face-to-face deliberation: Who? Why? What? With what effects? New Media & Society, 14(3), 363–383. https://doi.org/10.1177/1461444811413191

Bail

(2022). Breaking the social media prism: How to make our platforms less polarizing. Princeton University Press.

Bayram

A. B.

(2020). Measuring common knowledge: Latent semantic analysis, linguistic synchrony, and common knowledge in international relations. International Relations, 34(2), 180–203. https://doi.org/10.1177/0047117819871996

Bejan

T. M.

(2017). Mere civility: Disagreement and the limits of toleration. Harvard University Press.

10.

Bertram

Randazzo

Alabi

Levenson

Doucette

J. T.

Barbosa

(2016). Strong correlations between empathy, emotional intelligence, and personality traits among podiatric medical students: A cross-sectional study. Education for Health, 29(3), 186–194. https://doi.org/10.4103/1357-6283.204224

11.

Blevins

D. P.

Stackhouse

M. R.

Dionne

S. D.

(2022). Righting the balance: Understanding introverts (and extraverts) in the workplace. International Journal of Management Reviews, 24(1), 78–98. https://doi.org/10.1111/ijmr.12268

12.

Breil

S. M.

Geukes

Wilson

R. E.

Nestler

Vazire

Back

M. D.

(2019). Zooming into real-life extraversion – how personality and situation shape sociability in social interactions. Collabra: Psychology, 5(1), 7. https://doi.org/10.1525/collabra.170

13.

Bright

(2016). Explaining the emergence of echo chambers on social media: The role of ideology and extremism. arXiv preprint arXiv:1609.05003.

14.

Chambers

J. R.

Baron

R. S.

Inman

M. L.

(2006). Misperceptions in intergroup conflict: Disagreeing about what we disagree about. Psychological Science, 17(1), 38–45. https://doi.org/10.1111/j.1467-9280.2005.01662.x

15.

Chavez

L. R.

Campos

Corona

Sanchez

Ruiz

C. B.

(2019). Words hurt: Political rhetoric, emotions/affect, and psychological well-being among Mexican-origin youth. Social Science & Medicine, 228, 240–251. https://doi.org/10.1016/j.socscimed.2019.03.008

16.

Cinelli

De Francisci Morales

Galeazzi

Quattrociocchi

Starnini

(2021). The echo chamber effect on social media. Proceedings of the National Academy of Sciences, 118(9), e2023301118. https://doi.org/10.1073/pnas.2023301118

17.

Clark

H. H.

Brennan

S. E.

(1991). Grounding in communication. In Resnick

L. B.

Levine

J. M.

Teasley

S. D.

(Eds.), Perspectives on socially shared cognition (pp. 127–149). American Psychological Association.

18.

Cohen

Strayer

(1996). Empathy in conduct-disordered and comparison youth. Developmental Psychology, 32(6), 988–998. https://doi.org/10.1037/0012-1649.32.6.988

19.

Decety

Cowell

J. M.

(2014). The complex relation between morality and empathy. Trends in Cognitive Sciences, 18(7), 337–339. https://doi.org/10.1016/j.tics.2014.04.008

20.

Dunn

(2019). Public highly critical of state of political discourse in the U.S. Pew Research Center. https://www.pewresearch.org/politics/2019/06/19/public-highly- critical-of-state-of-political-discourse-in-the-u-s/

21.

Eaton

L. G.

Funder

D. C.

(2003). The creation and consequences of the social world: An interactional analysis of extraversion. European Journal of Personality, 17(5), 375–395. https://doi.org/10.1002/per.477

22.

Echterhoff

Higgins

E. T.

Levine

J. M.

(2009). Shared reality: Experiencing commonality with others’ inner states about the world. Perspectives on Psychological Science, 4(5), 496–521. https://doi.org/10.1111/j.1745-6924.2009.01161.x

23.

Ensley

M. D.

Pearson

A. W.

Amason

A. C.

(2002). Understanding the dynamics of new venture top management teams: Cohesion, conflict, and new venture performance. Journal of Business Venturing, 17(4), 365–386. https://doi.org/10.1016/S0883-9026(00)00065-3

24.

Fleeson

Gallagher

(2009). The implications of Big Five standing for the distribution of trait manifestation in behavior: Fifteen experience-sampling studies and a meta- analysis. Journal of Personality and Social Psychology, 97(6), 1097–1114. https://doi.org/10.1037/a0016786

25.

Gesn

P. R.

Ickes

(1999). The development of meaning contexts for empathic accuracy: Channel and sequence effects. Journal of Personality and Social Psychology, 77(4), 746–761. https://doi.org/10.1037/0022-3514.77.4.746

26.

Gonzales

A. L.

Hancock

J. T.

Pennebaker

J. W.

(2010). Language style matching as a predictor of social dynamics in small groups. Communication Research, 37(1), 3–19. https://doi.org/10.1177/0093650209351468

27.

Haas

B. W.

Brook

Remillard

Ishak

Anderson

I. W.

Filkowski

M. M.

(2015). I know how you feel: The warm-altruistic personality profile and the empathic brain. PLOS ONE, 10(3), e0120639. https://doi.org/10.1371/journal.pone.0120639

28.

Hall

(1993). Encoding, decoding. The Cultural Studies Reader, 4, 90–103.

29.

Hartmann

Tanis

(2013). Examining the hostile media effect as an intergroup phenomenon: The role of ingroup identification and status. Journal of Communication, 63(3), 535–555. https://doi.org/10.1111/jcom.12031

30.

Higgins

E. T.

Rossignac-Milon

Echterhoff

(2021). Shared reality: From sharing-is- believing to merging minds. Current Directions in Psychological Science, 30(2), 103–110. https://doi.org/10.1177/0963721421992027

31.

Hinds

P. J.

Bailey

D. E.

(2003). Out of sight, out of sync: Understanding conflict in distributed teams. Organization Science, 14(6), 615–632. https://doi.org/10.1287/orsc.14.6.615.24872

32.

Ickes

(2010). Everyday mind reading: Understanding what other people think and feel. Prometheus Books.

33.

Ireland

M. E.

Pennebaker

J. W.

(2010). Language style matching in writing: Synchrony in essays, correspondence, and poetry. Journal of Personality and Social Psychology, 99(3), 549–571. https://doi.org/10.1037/a0020386

34.

Jensen

(2016). Personality traits and nonverbal communication patterns. International Journal of Social Science Studies, 4(5), 57–70. https://doi.org/10.11114/ijsss.v4i5.1451

35.

Jolliffe

Farrington

D. P.

(2006). Examining the relationship between low empathy and bullying. Aggressive Behavior: Official Journal of the International Society for Research on Aggression, 32(6), 540–550. https://doi.org/10.1002/ab.20154

36.

Kecskes

Zhang

(2009). Activating, seeking, and creating common ground: A socio- cognitive approach. Pragmatics & Cognition, 17(2), 331–355. https://doi.org/10.1075/p&c.17.2.06kec

37.

Koerner

A. F.

Fitzpatrick

M. A.

(2002). You never leave your family in a fight: The impact of family of origin on conflict-behavior in romantic relationships. Communication Studies, 53(3), 234–251. https://doi.org/10.1080/10510970209388588

38.

Krauss

R. M.

(2002). The psychology of verbal communication. In International encyclopaedia of the social and behavioral sciences (pp. 16161–16165). Elsevier.

39.

Landauer

T. K.

Dumais

S. T.

(1997). A solution to Plato's problem: The latent semantic analysis theory of acquisition, induction, and representation of knowledge. Psychological Review, 104(2), 211–240. https://doi.org/10.1037/0033-295X.104.2.211

40.

Landauer

T. K.

Foltz

P. W.

Laham

(1998). An introduction to latent semantic analysis. Discourse Processes, 25(2-3), 259–284. https://doi.org/10.1080/01638539809545028

41.

Leith

K. P.

Baumeister

R. F.

(1998). Empathy, shame, guilt, and narratives of interpersonal conflicts: Guilt-prone people are better at perspective taking. Journal of Personality, 66(1), 1–37. https://doi.org/10.1111/1467-6494.00001

42.

Lucas

R. E.

Diener

Grob

Suh

E. M.

Shao

(2000). Cross-cultural evidence for the fundamental features of extraversion. Journal of Personality and Social Psychology, 79(3), 452–468. https://doi.org/10.1037/0022-3514.79.3.452

43.

McDaniel

B. T.

Coyne

S. M.

(2016). “Technoference”: The interference of technology in couple relationships and implications for women’s personal and relational well-being. Psychology of Popular Media Culture, 5(1), 85–98. https://doi.org/10.1037/ppm0000065

44.

Mehl

M. R.

Gosling

S. D.

Pennebaker

J. W.

(2006). Personality in its natural habitat: Manifestations and implicit folk theories of personality in daily life. Journal of Personality and Social Psychology, 90(5), 862–877. https://doi.org/10.1037/0022-3514.90.5.862

45.

Morganti

(2008). What intersubjectivity affords: Paving the way for a dialogue between cognitive science, social cognition and neuroscience. In F. Morganti, A. Carassa, & G. Riva (Eds.), Enacting intersubjectivity: A cognitive and social perspective on the study of interactions (pp. 3–14). IOS Press.

46.

Motley

M. T.

(1986). Consciousness and intentionality in communication: A preliminary model and methodological approaches. Western Journal of Communication (Includes Communication Reports), 50(1), 3–23. https://doi.org/10.1080/10570318609374210

47.

Pang

Zhou

Gao

Han

(2023). Are women more empathetic than men? Questionnaire and EEG estimations of sex/gender differences in empathic ability. Social Cognitive and Affective Neuroscience, 18(1), nsad008. https://doi.org/10.1093/scan/nsad008

48.

Pennebaker

J. W.

Graybeal

(2001). Patterns of natural language use: Disclosure, personality, and social integration. Current Directions in Psychological Science, 10(3), 90–93. https://doi.org/10.1111/1467-8721.00123

49.

Peters

Verboon

(2023). rosetta: Parallel use of statistical packages in teaching. R package version 0.3.12. https://CRAN.R-project.org/package=rosetta

50.

Pew Research Center. (2019). Partisan antipathy: More intense, more personal. https://www.pewresearch.org/politics/2019/10/10/partisan-antipathy-more-intense-more- personal/

51.

Pew Research Center. (2022). America's abortion quandary. https://www.pewresearch.org/religion/2022/05/06/americas-abortion-quandary

52.

Pew Research Center. (2025, June 12). Public opinion on abortion: Views on abortion, 1995-2024. https://www.pewresearch.org/religion/fact-sheet/public-opinion-on-abortion/

53.

Placiński

Żywiczyński

(2023). Modality effect in interactive alignment: Differences between spoken and text-based conversation. Lingua. International review of general linguistics. Revue Internationale De linguistique Generale, 293, 103592. https://doi.org/10.1016/j.lingua.2023.103592

54.

Rammstedt

John

O. P.

(2007). Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German. Journal of Research in Personality, 41(1), 203–212. https://doi.org/10.1016/j.jrp.2006.02.001

55.

R Core Team. (2024). R: A language and environment for statistical computing (Version 4.4.0) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/

56.

Rossignac-Milon

Bolger

Zee

K. S.

Boothby

E. J.

Higgins

E. T.

(2021). Merged minds: Generalized shared reality in dyadic relationships. Journal of Personality and Social Psychology, 120(4), 882–911. https://doi.org/10.1037/pspi0000266

57.

Rossini

(2022). Beyond incivility: Understanding patterns of uncivil and intolerant discourse in online political talk. Communication Research, 49(3), 399–425. https://doi.org/10.1177/0093650220921314

58.

Sacristan

Royle

Galpin

(2023). Texting and the brain: The time-course of social brain activation in face-to-face versus text-based computer-mediated-communication. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 17(4), Article 3. https://doi.org/10.5817/CP2023-4-3

59.

Saveski

Gillani

Yuan

Vijayaraghavan

Roy

(2022, May). Perspective- taking to reduce affective polarization on social media. In Proceedings of the international AAAI conference on web and social media (Vol. 16, pp. 885–895).

60.

Schumann

Zaki

Dweck

C. S.

(2014). Addressing the empathy deficit: Beliefs about the malleability of empathy predict effortful responses when empathy is challenging. Journal of Personality and Social Psychology, 107(3), 475–493. https://doi.org/10.1037/a0036738

61.

Sharma

Saha

Ernala

S. K.

Ghoshal

De Choudhury

(2017, October). Analyzing ideological discourse on social media: A case study of the abortion debate. In Proceedings of the 2017 International Conference of the Computational Social Science Society of the Americas (pp. 1–8). Association for Computing Machinery.

62.

Sherman

R. A.

Rauthmann

J. F.

Brown

N. A.

Serfass

D. G.

Jones

A. B.

(2015). The independent effects of personality and situations on real-time expressions of behavior and emotion. Journal of Personality and Social Psychology, 109(5), 872–888. https://doi.org/10.1037/pspp0000036

63.

Skytte

(2021). Dimensions of elite partisan polarization: Disentangling the effects of incivility and issue polarization. British Journal of Political Science, 51(4), 1457–1475. https://doi.org/10.1017/S0007123419000760

64.

Spreng

R. N.

McKinnon

M. C.

Mar

R. A.

Levine

(2009). The Toronto Empathy Questionnaire: Scale development and initial validation of a factor-analytic solution to multiple empathy measures. Journal of Personality Assessment, 91(1), 62–71. https://doi.org/10.1080/00223890802484381

65.

L. Y.

Cacciatore

M. A.

Scheufele

D. A.

Brossard

(2018). Online political discourse: Exploring differences in incivility in the comments section of news websites. Social Media and Society, 4(1), 1–11. https://doi.org/10.1177/1461444818757205

66.

V. P.

Babcock

M. J.

Ickes

(2017). Developing latent semantic similarity in initial, unstructured interactions: The words may be all you need. Journal of Language and Social Psychology, 36(2), 143–166. https://doi.org/10.1177/0261927X16638386

67.

V. P.

Ickes

(2020). Latent semantic similarity in initial computer-mediated interactions: Moderating effects of time, extraversion, and gender composition. International Journal of Interactive Communication Systems and Technologies, 10(1), 51–71. https://doi.org/10.4018/IJICST.2020010104

68.

Ta-Johnson, V. P., & Swafford, I. M. (2026). Empathy mediates the link between extraversion and latent semantic similarity in computer-mediated discussions of a politically contentious topic: Evidence from abortion-related discourse [Dataset]. Open Science Framework. https://osf.io/ucvbk/overview

69.

Vaccari

Valeriani

(2021). Outside the bubble: Social media and political participation in Western democracies. Oxford University Press. https://doi.org/10.1093/oso/9780190858476.001.0001

70.

Van Elsas

Fiselier

(2024). Conflict or choice? The differential effects of elite incivility and ideological polarization on political support. Acta Politica, 59(3), 589–618. https://doi.org/10.1057/s41269-023-00288-5

71.

Van Langen

M. A.

Wissink

I. B.

van Vugt

E. S.

Van der Stouwe

Stams

G. J. J.

(2014). The relation between empathy and offending: A meta-analysis. Aggression and Violent Behavior, 19(2), 179–189. https://doi.org/10.1016/j.avb.2014.02.003

72.

Vrana

S. R.

Vrana

D. T.

Penner

L. A.

Eggly

Slatcher

R. B.

Hagiwara

(2018). Latent Semantic Analysis: A new measure of patient-physician communication. Social Science & Medicine, 198, 22–26. https://doi.org/10.1016/j.socscimed.2017.12.021

73.

Watson

Stasik

S. M.

Ellickson-Larew

Stanton

(2015). Extraversion and psychopathology: A facet-level analysis. Journal of Abnormal Psychology, 124(2), 432–446. https://doi.org/10.1037/abn0000051

74.

Webster

S. W.

Connors

E. C.

Sinclair

(2022). The social consequences of political anger. The Journal of Politics, 84(3), 1292–1305. https://doi.org/10.1086/718979

75.

Willard

N. E.

(2007). Cyberbullying and cyberthreats: Responding to the challenge of online social aggression, threats, and distress. Research Press.

76.

Wilt

Revelle

(2009). Extraversion. In Leary

M. R.

Hoyle

R. H.

(Eds.), Handbook of individual differences in social behavior (pp. 27–45). The Guilford Press.

77.

Wojcieszak

Bimber

Feldman

Stroud

N. J.

(2015). Partisan news and political participation: Exploring mediated relationships. Political Communication, 33(2), 241–260. https://doi.org/10.1080/10584609.2015.1051608

78.

Zhang

Kong

(2023). Sex difference in trait empathy is encoded in the human anterior insula. Cerebral Cortex, 33(9), 5055–5065. https://doi.org/10.1093/cercor/bhac398

79.

Zaki

(2014). Empathy: A motivated account. Psychological Bulletin, 140(6), 1608–1647. https://doi.org/10.1037/a0037679

80.

Zohoorian

Zeraatpishe

Khorrami

(2022). Willingness to communicate, big five personality traits, and empathy: How are they related?. Canadian Journal of Educational and Social Studies, 2(5), 17–27. https://doi.org/10.53103/cjess.v2i5.69

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

0.19 MB