Blinded by guilt: Short-term relational focus and lying

Abstract

Guilt has often been conceptualized as a moral emotion because it stimulates transgressors to restore a damaged relationship. However, not much has been investigated about when the guilt-induced relational-promoting goal conflicts with certain moral norms (for instance, honesty). In this paper, we demonstrate that guilt increases the transgressors' lying intentions in situations where honesty is anticipated to incur immediate relational harm with the victim and that this effect is driven by the transgressors' short-term relational focus. Across three experiments (two preregistered), we provide robust evidence for our prediction. Study 1 examined the effect of guilt on inflating the feedback to the victim, measured relational focus and the focus on the instrumental value of honesty (i.e., long-term benefits brought by honesty such as career growth) as simultaneous mechanisms, and contrasted the effect of guilt on the lying intention with that of shame. Study 2 better measured the short-termism in the relational focus and examined a situation wherein guilty participants could lie to a third party for the benefit of the victim. Study 3 provided additional support for our proposed mechanism by showing that deliberation on the consequences of not being honest (i.e., indirectly increasing participants' focus on the long-term instrumental value of honesty) decreases people's lying intentions. These results contribute to our understanding of the complex interpersonal nature of guilt and, specifically, its temporal focus.

Introduction

Elicited by interpersonal harm, guilt stimulates transgressors towards relationship promoting behaviors such as cooperation, compensation, and reparation (e.g., Baumeister, Stillwell, & Heatherton, 1994; Cryder, Springer, & Morewedge, 2012; Graton, Ric, & Gonzalez, 2016; Hoffman, 1994; Tangney, Stuewig, & Mashek, 2007). In fact, guilt is also referred to as a moral emotion because it improves the welfare of individuals and the society as a whole (Haidt, 2003).

However, the excessive attention on interpersonal relationships brought by guilty feelings can be problematic under certain circumstances. For example, imagine one day a colleague whom you have accidentally harmed and hence felt guilty towards, came to you for suggestions on their project presentation, and suppose in your honest opinion the presentation was poorly done. In such a scenario, you need to choose between giving honest (and unpleasant) feedback that may cause immediate relational harm but may improve the presentation in the long run or giving false praise that violates the moral norm of honesty.

To our knowledge, there is limited, if any, research investigating how guilty transgressors will behave when the goal of relationship promotion is incompatible with certain moral norms. In the current study, we demonstrate that guilt can evoke violation of the highly valued and widely accepted moral norm of honesty, especially when honesty is anticipated to incur immediate relational harm with the victim. However, this effect could be attenuated by directing the guilty transgressors' attention to the instrumental value of honesty (i.e., potential long-term benefits of honesty such as meaningful growth, understanding, or change: Levine, 2021). With these findings, we broaden our understanding of the interpersonal function of guilt and its temporal focus.

As a self-conscious negative emotion, guilt often arises from transgressions such as interpersonal neglect, unfilled obligations, and selfish actions in social interactions where the transgressor has harmed another person and violated important norms (e.g., Baumeister et al., 1994; Lewis, 1971). Guilt is also referred to as a “social emotion” because it is “connected with (real or imagined) social interactions” (Barrett, 1995, p.25), and it is experienced more frequently in communal and close relationships (e.g., Baumeister, Stillwell, & Heatherton, 1995).

Accordingly, guilt serves to enhance one's interpersonal relationships and studies have consistently found that guilt is adaptive in its role in relationship maintenance (Ausubel, 1955; O'Connor, Berry, & Weiss, 1999). For instance, guilt stimulates transgressors to take reparative actions such as confession, apology, remorse, and compensation to undo the harm (Baumeister et al., 1994; Lewis, 1971; Lindsay-Hartz, 1984; Tangney, 1991). Guilt also enhances compliance and cooperation (de Hooge, Zeelenberg, & Breugelmans, 2007; Freedman & Fraser, 1966; Ketelaar & Tung Au, 2003) and fosters other-oriented empathy and perspective-taking (Hoffman, 1994; Yang, Yang, & Chiou, 2010). The goal to repair a damaged relationship can be strong enough such that a guilty transgressor may compensate the victim at the expense of a third-party's best interests (de Hooge, Nelissen, Breugelmans, & Zeelenberg, 2011).

Consistent with the previous research, we define lying as when a communicator intentionally tries to mislead the target and this could be about the objective truth or one's subjective feelings and beliefs that are independent of reality (Boles, Croson, & Murnighan, 2000; DePaulo, Kirkendol, Kashy, Wyer, & Epstein, 1996; Levine & Cohen, 2018; Levine & Schweitzer, 2015).

Lying to a partner is sometimes driven by a relational focus. For example, DePaulo et al. (1996) found that people can lie to prevent conflict, and Ennis, Vrij, and Chance (2008) found that lies to prevent interpersonal conflicts are told more in close relationships. Relatedly, people tend to believe that honesty can sometimes damage the relationship. For instance, people anticipate less liking from poorly-performing recipients if asked to provide honest feedback (Fisher, 1979) and avoid complete self-disclosure in social interactions when complete self-disclosure can hurt others (Rosenfeld, 1979). People are also reluctant to deliver an honest yet hurtful message to a partner due to concerns of being perceived negatively by their partner (Bond & Anderson, 1987). Interestingly, Levine and Cohen (2018) demonstrated that people mispredicted the relational harm brought by honesty such that honesty did less harm than what individuals expected. Irrespective of whether unpleasant honesty is indeed harmful for relationships or not, previous research suggests that lying commonly occurs with the expectation of enhancing interpersonal relationships (e.g., Metts, 1989).

In addition, lying to a third party is expected to protect the relationship between the liar and their partners. For example, Kouchaki and Kray (2018) found that females were more likely to lie to a negotiator when advocating for a partner due to greater relational sensitivity, and Zhu et al. (2020) found that gratitude motivated individuals to lie to third parties t to protect their partner and build a relationship.

Although lying might sometimes help avoid immediate relational harm, it may lead to lost benefits that could have resulted from honesty. For example, giving overly positive feedback to avoid interpersonal tension can inhibit the partner from performance improvement (e.g., Ellis, Mendel, & Aloni-Zohar, 2009; Jampol & Zayas, 2021; Lupoli, Jampol, & Oveis, 2017), and false praise may even lead to hubris that is detrimental for the partner (e.g., Hayward & Hambrick, 1997). Lying to a third party may also jeopardize the third party's best interests when making a decision requires them to obtain truthful information (e.g., Kouchaki & Kray, 2018). In addition, the detection of lies can further worsen, rather than improve, the relationship quality in the long run (e.g., Cole, 2001; Lupoli, Levine, & Greenberg, 2018). As such, situations of unpleasant honesty involve a tradeoff between lying to avoid immediate relational harm and the long-term instrumental benefits associated with honesty. Our framework is consistent with the recent work from Levine (2021), who has provided a parsimonious theoretical account of the acceptability of lying in everyday conversations, such that individuals are more likely to accept lies when the degree of the immediate harm of honesty (e.g., the harm at the moment of communication) is high and the instrumental value of honesty (e.g., the long-term benefits of honesty) is low.

Given the relational focus of guilt, one could conjecture that guilt would decrease lying because it may cause long-term relational damage. The detection that one has been lied to is linked to deleterious interpersonal outcomes such as decreased liking (Tyler, Feldman, & Reichert, 2006), trust (Schweitzer, Hershey, & Bradlow, 2006), and relational commitment (Cole, 2001). Lying detection is often more devastating in close relationships because of high expectations from each other in such relationships (Miller, Mongeau, & Sleight, 1986). Even for the lies that are out of the liar's good intentions, studies have shown that recipients held a strong distaste towards such lies (Lupoli et al., 2018). Since the focus of guilt-induced behavior is to enhance one's relationships, one could predict guilt to prevent lying.

However, in situations where honesty may create immediate relational harm, we suggest that guilt heightens short-term relational focus which, in turn, would increase lying. Functional accounts of emotions indicate that emotions direct people's attention to the problems at hand and urge people to take actions (e.g., Hajcak et al., 2007; Keltner & Gross, 1999; Keltner & Haidt, 1999; Keltner & Kring, 1998; Oatley & Jenkins, 1992). Keltner and Haidt (1999) also suggested that one basic function of emotion is to inform people about their social conditions “typically needing to be acted upon and changed” (p.509). In other words, emotions can create a sense of urgency to focus on the problem that needs to be particularly addressed based on the specific motivational functions of each emotion (Zeelenberg, Nelissen, Breugelmans, & Pieters, 2008). Accordingly, considerable evidence shows that emotions—especially the heightened ones—often make people rush into actions that are short-term focused (e.g., Cyders & Smith, 2008). For example, Lupoli et al. (2017) found that compassion increased participants' lying to the target to avoid immediate emotional harm, even though honest feedback could improve the target's chance to win a prize in a writing game. Gray (1999) found that people experiencing negative emotions preferred short-term benefits and forwent long-term consequences. Furthermore, Fischer and Roseman (2007) found that anger prompted attacking through a short-term focus on coercion while contempt prompted social exclusion through a short-term focus on derogation. In a similar vein, given that guilt engenders attention to the damaged relationship, guilty transgressors would feel a sense of urgency to repair it. Therefore, we predict that when honesty can evoke immediate relational harm with the victim, guilt should increase lying.

Although not explicitly stated, previous studies provided evidence that guilt might make individuals short-sighted with a preoccupation with relational focus, and this is because guilt can direct individuals' attention to reparatory means (Graton & Ric, 2017). In their comprehensive review, Graton and Mailliez (2019) suggested that upon feeling guilty, transgressors may allocate specific attention to immediate reparation that predicted further behaviors. This proposition is corroborated by empirical findings. For example, Ketelaar and Tung Au (2003) found that guilty individuals were more cooperative only during the first ten trials from a total of forty trials in repeated social bargaining games and de Hooge et al. (2007) replicated this finding and further termed the effect of guilt on cooperation as short-termed. Similarly, related literature found that guilt motivated approach tendencies when an opportunity for reparation presented itself (Amodio, Devine, & Harmon-Jones, 2007) and that once a compensation had been made to the victim, the guilty transgressor no longer exerted restoration efforts (de Hooge, 2012; de Hooge, Nelissen, et al., 2011). These results suggest that guilty transgressors will seize the proximal opportunity to repair the relationship, but they are unlikely to continue doing so.

Separately, in a consumer behavior context, Han, Duhachek, and Agrawal (2014) explored the relationship between guilt and psychological distance using construal level theory (e.g., Liberman & Trope, 1998) and found that guilt elicited lower level (i.e., more concrete) construals, which is related to a heightened focus on proximal vis-à-vis distal outcomes (e.g., Bar-Anan, Liberman, & Trope, 2006). Recently, Pounders, Royne, and Lee (2019) supported this notion by showing that guilt appeals were more persuasive when paired with a proximal frame due to higher processing fluency, which further suggested that guilt-laden individuals adopted lower-level construals.

Taken together, previous literature suggests that guilt can make individuals focus on short-term vis-à-vis more distant goals. Therefore, in situations where honesty can bring immediate relational harm, we predict that guilty transgressors are more likely to lie due to their short-term relational focus.

In three experiments, we found that in situations where honesty is anticipated to cause immediate relational harm, guilty participants demonstrated higher lying intentions than the control group. Study 1 examined the effect of guilt on inflating feedback, wherein we measured relational focus and the focus on the instrumental value of honesty as two simultaneous mechanisms and contrasted the effect of guilt on lying intentions with that of shame (another closely related negative emotion). Study 2 extended our investigation by capturing the short-termism in the measure of relational focus and by examining a situation wherein guilty participants could lie to a third party. In Study 3, we provided additional support for our proposed mechanism by asking participants to deliberate on the consequences of not being honest (thereby, indirectly increasing participants' focus on the instrumental value of honesty).

We pre-registered Study 1 and Study 3 and reported all measures, manipulations, exclusions, and sensitivity analyses for all studies. Vignettes were employed in each study to strictly control for the nature of the relationship between the transgressor and the victim, as previous literature has found that the attributes of a relationship could influence transgressors' experience of guilt (Baumeister et al., 1995; Nelissen, 2014). As such, we were only able to measure lying intentions (vis-à-vis actual lying behaviors) based on hypothetical scenarios. However, we believe it is unlikely to impact the validity of our results and we provide a more detailed explanation in the General Discussion section. All data and materials can be found in OSF (https://osf.io/kq67n/?view_only=091686b70a64491384e3965c88c0d8b9).