Emma Levine

Why do people avoid having important but difficult conversations? The dominant explanation is that people are overly pessimistic about the outcomes of conversations— overestimating costs and underestimating benefits. However, anecdotal evidence suggests people still fail to initiate conversations they expect to be more beneficial than costly. We address this puzzle by testing a self-control account of difficult conversations. Like other classic self-control conflicts (e.g., going to the gym, eating healthily), we argue that engaging in difficult conversations requires people to overcome proximal, concrete costs (e.g., feeling uncomfortable, hurting someone’s feelings in the moment) in pursuit of distal, abstract benefits (e.g., helping someone grow). Across four pilot studies and six experiments (N = 2,442), we provide three key pieces of support for our account. First, people perceive difficult conversations as having intertemporal trade-offs, such that the costs precede the benefits. Second, people’s decisions to initiate difficult conversations are sensitive to time, such that people are more willing to have difficult conversations in the future than the present (i.e., they delay). Importantly, this effect is specific to conversations that are expected to have both costs and benefits, suggesting that people do not generally delay all conversations. This delay is mediated by differential weighing of the costs and benefits in addition to lower anticipated costs in the future (vs. the present). Third, people are more willing to take up costly commitment devices to motivate themselves to initiate difficult conversations than to take up equivalently costly gambles. This suggests that people are aware of these dynamics. Overall, we provide a new theoretical lens for understanding difficult conversations and point to new interventions to prompt them.

Decision-makers must often choose whether to commit necessary evils (i.e., cause harm to produce a benefit), such as allocating punishments or redistributing resources. Existing moral psychological work on instrumental harm suggests that people should react poorly to necessary evils and condemn those who commit them. However, four preregistered studies (total N = 1,622), using both survey experiments and incentive-compatible games, reveal that targets judge decision-makers who commit a broad range of everyday necessary evils more positively than decision-makers expect. Decision-makers overestimate how much emotional harm targets experience and underestimate how much instrumental value targets recognize. As a result, decision-makers overestimate the reputational costs of performing necessary evils, which can serve as a barrier to performing them. This research highlights the theoretical importance of studying social-moral phenomena (such as instrumental harm) in everyday contexts and provides practical insight into how to encourage decision-makers to perform societally beneficial, but seemingly harmful actions.

In this work, we introduce the Experience-Outcome Model of honest conversation. Our model advances three key ideas. First, a communicator’s decision to initiate a conversation is influenced by their expected experience during a conversation, in addition to the expected outcomes following the conversation. Second, these same factors influence a communicator’s decision to speak truthfully within a conversation, conditional on initiation. Third, when communicators expect the experience of a conversation to be negative but the outcomes of a conversation to be positive, this leads to self-regulatory conflict. After introducing the Experience-Outcome Model, we explain its implications for difficult conversations—conversations that are aversive because they can cause interpersonal harm. The Experience-Outcome Model explains why communicators fail to initiate difficult conversations and why they communicate dishonestly within them, even when communicators expect the conversations to lead to positive outcomes: The negative experience associated with harm aversion creates self-regulatory conflict. Drawing from the literature on self-regulation, we develop six key propositions that explain how psychological distance influences the experience of harm aversion within difficult conversations and, therefore, a communicator’s likelihood of overcoming this conflict. Our model deepens the current understanding of the psychological dynamics underlying honesty and social engagement. By applying the Experience-Outcome Model to difficult conversations, we provide insight into the psychology of self-regulation and everyday harm as well as provide practical guidance on managing, mentoring, and promoting others’ growth.

Organizational brokers, such as producers, diplomats, and middle managers, connect people within and across organizations. Trust in brokers promotes their effectiveness in connecting unfamiliar counterparts and transferring information between disconnected parties. Often, however, the very conditions that create the need for brokering, such as the existence of structural holes, undermine trust in brokers. We introduce an organizing framework to explain how fundamental features of brokering processes shape the perceived trustworthiness of brokers. Specifically, we describe how individuals make inferences about brokers’ ability, benevolence, and integrity based on the horizontal and vertical distances they bridge, the extent to which their brokering involves in-role versus extra-role behavior, and the extent to which they facilitate indirect versus direct exchanges between counterparts. Our model links situational features of brokering processes with established antecedents of interpersonal trust, thereby identifying challenges and opportunities for brokers and organizations.

Honesty, defined as freedom from fraud or deception, is widely valued in many aspects of life, from personal relationships to professional settings. Yet acts of dishonesty remain widespread, including political and corporate scandals, misinformation, personal betrayal, and so on. Understanding honesty and the factors that influence it provides insights that are essential for fostering trust and combating corruption. In this review, we synthesize key findings from research on honesty, focusing on when people choose to be truthful or deceptive. We argue that although much is known about honesty in isolated, low-risk contexts, an urgent need exists to study honesty in more complex, realistic settings, such as those involving interpersonal relationships, potential sanctions, or group influences. Our proposed framework highlights understudied contexts and encourages future studies to explore settings where enforcement and social dynamics play a significant role in decision-making. To do so, we point out 66 open research questions that we find most promising to explore. By integrating insights from multiple disciplines, we aim to advance the understanding of honesty and provide a roadmap for research that can inform policies and interventions to promote integrity in society.

People believe that some lies are ethical, while also claiming that "honesty is the best policy." In this article, we introduce a theory to explain this apparent inconsistency. Even though people view prosocial lies as ethical, they believe it is more important-and more moral-to avoid harmful lies than to allow prosocial lies. Unconditional honesty (simply telling the truth, without finding out how honesty will affect others) is therefore seen as ethical because it prevents the most unethical actions (i.e., harmful lies) from occurring, even though it does not optimize every moral decision. We test this theory across five focal experiments and 10 supplemental studies. Consistent with our account, we find that communicators who tell the truth without finding out how honesty will affect others are viewed as more ethical, and are trusted more, than communicators who look for information about the social consequences of honesty before communicating. However, the moral preference for unconditional honesty attenuates when it is certain that looking for more information will not lead to harmful lies. Overall, this research provides a holistic understanding of how people think about honesty and suggests that moral rules are not valued because people believe all rule violations are wrong, but rather, because they believe some violations must be avoided entirely. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Despite the well-documented costs of word-deed misalignment, hypocrisy permeates our personal, professional, and political lives. Why? We explore one potential explanation: the costs of moral flexibility can outweigh the costs of hypocrisy, making hypocritical moral absolutism a preferred social strategy to admissions of moral nuance. We study this phenomenon in the context of honesty. Across six studies (total N = 3545), we find that communicators who take flexible honesty stances ("It is sometimes okay to lie") that align with their behavior are penalized more than hypocritical communicators who take absolute honesty stances ("It is never okay to lie") that they fail to uphold. Although few people take absolute stances against deception themselves, they are more trusting of communicators who take absolute honesty stances, relative to flexible honesty stances, because they perceive absolute stances as reliable signals of communicators' likelihood of engaging in future honesty, regardless of inconsistent behavior. Importantly, communicators-including U.S. government officials-also anticipate the costs of flexibility. This research deepens our understanding of the psychology of honesty and helps explain the persistence of hypocrisy in our social world. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Although consistency has long been positioned as a cornerstone of trust, we explore situations in which consistency undermines trust. Across one pilot study and seven preregistered experiments (total N = 3,000), we study trust in authority figures (e.g., police officers, managers) who either consistently punish offenders according to codified rules (e.g., laws, policies) or who exercise discretion by occasionally deviating from rules. We find that people are more likely to trust authority figures that exercise discretion rather than consistency to the extent that discretion signals benevolence. Specifically, people reward authority figures who use discretion leniently (rather than punitively) and apply it thoughtfully (rather than arbitrarily). Moreover, only certain cases are deemed fit for lenient discretion. When discretion is perceived to be motivated by favoritism because it is applied to close others, or when the basis for discretion is unclear because there is little variance in cases of the crime being punished, discretion fails to signal benevolence and trustworthiness. This research has important implications for understanding trust, discretion, and the reputational consequences of punishment.

Targeted therapies have revolutionized the treatment for many patients with non-small cell lung cancer (NSCLC). Multiple new oral targeted therapies have been approved in the last decade; however, their overall efficacy may be reduced by poor adherence, treatment interruptions, or dose reductions due to adverse events. Most institutions lack standard monitoring protocols for toxicities from these targeted agents. This review describes important adverse events observed in clinical trials and reported by the U.S. Food and Drug Administration for both currently approved and upcoming promising therapies in the treatment of NSCLC. These agents cause a range of toxicities, including dermatologic, gastroenteric, pulmonary, and cardiac toxicities. This review proposes protocols for routine monitoring of these adverse events, both prior to initiation of therapy and while on treatment.

This page contains materials, data, and access to the R scripts for all studies included in the article "On being honest about dishonesty: The social costs of taking nuanced (but realistic) moral stances" (Journal of Personality and Social Psychology, 2023). Please note: © 2022, American Psychological Association. This paper is not the copy of record and may not exactly replicate the final, authoritative version of the article. Please do not copy or cite without authors' permission. The final article will be available, upon publication, via its DOI: 10.1037/pspa0000340