A Theory of Blame

AI contextual information shapes moral and aesthetic judgments of AI-generated visual art

Throughout history, art creation has been regarded as a uniquely human means to express original ideas, emotions, and experiences. However, as Generative Artificial Intelligence reshapes visual, aesthetic, legal, and economic culture, critical questions arise about the moral and aesthetic implications of AI-generated art. Despite the growing use of AI tools in art, the moral impact of AI involvement in the art creation process remains underexplored. Understanding moral judgments of AI-generated art is essential for assessing AI's impact on art and its alignment with ethical norms. Across three pre-registered experiments combining explicit and implicit paradigms with Bayesian modelling, we examined how information about AI systems influences moral and aesthetic judgments and whether human art is implicitly associated with positive attributes compared to AI-generated art. Experiment 1 revealed that factual information about AI backend processes reduced moral acceptability and aesthetic appeal in certain contexts, such as gaining financial incentives and art status. Experiment 2 showed that additional information about AI art's success had no clear impact on moral judgments. Experiment 3 demonstrated that an implicit association task did not reliably link human art with positive attributes and AI art with negative ones. These findings show that factual information about AI systems shapes judgments, while different information doses about AI art's success have limited moral impact. Additionally, implicit associations between human-made and AI-generated art are similar. This work enhances understanding of moral and aesthetic perceptions of AI-generated art, emphasizing the importance of examining human-AI interactions in an arts context, and their current and evolving societal implications.

Is an eye truly for an eye? Magnitude differences affect moral praise more than moral blame

Does the amount of perceived moral responsibility correspond to the magnitude of the act to the same degree regardless of whether the act is moral or immoral? In four experiments (N = 1617; all preregistered), we found that-when evaluating two agents who performed similar acts but with different magnitude-observers judged greater differences in their moral responsibility when those acts were moral than when they were immoral. That is, the same difference in magnitude had greater influence on perceived moral responsibility for moral acts compared to immoral acts. Furthermore, we also found that the asymmetry effect impacted perceivers' judgment of the moral character of the agent (Studies 2 and 3). Evaluating immoral (vs. moral) acts led participants to use a more affect-based (vs. reason-based) decision mode, which, in turn, led them to be more scope insensitive to the magnitude difference of the two acts (Study 3). Lastly, we showed that this asymmetry effect is moderated by the individual's concern with the relevant moral issue (Study 4). When perceivers care less about the issue (e.g., animal welfare), the asymmetry effect attenuates. These results together suggest that, when comparing the moral responsibility of different moral agents, magnitude of behavior matters more for positive than for negative acts.

The Double Standard of Ownership

Owners are often blamed when their property causes harm but might not receive corresponding praise when their property does good. This suggests a double standard of ownership, wherein owning property poses risks for moral blame that are not balanced with equal opportunities for credit. We investigated this possibility in three preregistered experiments on 746 US residents. Participants read vignettes where agentic property (e.g., animals, robots) produced bad or good outcomes, and judged whether owners and the property were morally responsible. With bad outcomes, participants assigned owners more blame than property (Experiments 1 and 2) or similar blame (Experiment 3). But with good outcomes, participants consistently assigned owners much less praise relative to their property. The first two experiments also examined if the double standard arises in two other relationships between authorities and subordinates; participants showed the double standard when assessing moral responsibility for parents and children, but not for employers and employees. Together, these findings point to a novel asymmetry in how owners are assigned responsibility.

Who is to blame for the 'problem' of teenage pregnancy? Narratives of blame in two South African communities

BACKGROUND

The framing of teenage pregnancy in social discourse influences the way in which adolescent girls and young women (AGYW) are treated, the extent to which they are supported, and to which they are able to engage with services and enact pregnancy prevention behaviours.

METHODS

Through the analysis of data from a qualitative study conducted in the South African communities of Newcastle in KwaZulu-Natal and Moretele in the North-West province, we explored narratives of blame for teenage pregnancy. Data derived from seventy-one in-depth interviews conducted with forty AGYW aged between 15 and 24, twenty-four parents/caregivers of AGYW, five service providers, ten school educators, and two other community members. Interpretation of data drew upon blame, attribution and framing theories.

FINDINGS

The overall framing of teenage pregnancy was overwhelmingly negative, with participants labelling it immoral and 'wrong'. Findings were arranged into key thematic areas that emerged in the data relating to blame for teenage pregnancy: the individual centred 'wrong-girl' and developmental discourses, in which blame was apportioned to AGYW for being immature, reckless, irresponsible, disobedient and greedy; the 'wrong-family' discourse in which blame was attributed to parents of AGYW for poor parenting and failing to adequately care for the adolescents in their responsibility; the 'wrong-men' discourse in which blame was attributed to men for luring AGYW into transactional sex, and for perpetrating gender-based violence; and lastly the 'wrong-society' discourse which attributed blame to contextual/structural factors such as poverty, a high volume of unregulated drinking establishments, and a lack of youth-friendly services.

CONCLUSIONS

Policies, interventions and programmes focusing on adolescents' sexual and reproductive health need to carefully consider the framing and narratives of blame and responsibility. There is an urgent need to shift away from the individualised moralistic shaming of pregnant AGYW, towards a recognition of a complex interplay of multilevel factors that enable or constrain AGYW's agency. Policies and programmes should focus on both providing sufficient support to AGYW, while also addressing structural factors and harmful narratives and thus create the conditions necessary to enable AGYW to enact safe, informed, agentic and responsible decisions and behaviours about their own sexual and reproductive health.

The impact of action descriptions on attribution of moral responsibility towards robots

In the era of renewed fascination with AI and robotics, one needs to address questions related to their societal impact, particularly in terms of moral responsibility and intentionality. In seven vignette-based experiments we investigated whether the consequences of a robot or human's actions, influenced participant's intentionality and moral responsibility ratings. For the robot, when the vignettes contained mentalistic descriptions, moral responsibility ratings were higher for negative actions consequences than positive action consequences, however, there was no difference in intentionality ratings. Whereas, for the human, both moral responsibility and intentionality ratings were higher for negative action consequences. Once the mentalistic descriptions were removed from the vignettes and the moral responsibility question was clarified, we found a reversed asymmetry. For both robots and humans, participants attributed more intentionality and praise, for positive action consequences than negative action consequences. We suggest that this reversal could be due to people defaulting to charitable explanations, when explicit references to culpable mental states are removed from the vignettes.

On the Perception of Moral Standing to Blame

Is everyone equally justified in blaming another's moral transgression? Across five studies (four pre-registered; total N = 1,316 American participants), we investigated the perception of moral standing to blame-the appropriateness and legitimacy for someone to blame a moral wrongdoing. We propose and provide evidence for a moral commitment hypothesis-a blamer is perceived to have low moral standing to blame a moral transgressor if the blamer demonstrates weak commitment to that moral rule. As hypothesized, we found that when blamers did not have the chance or relevant experience to demonstrate good commitment to a moral rule, participants generally believed that they had high moral standing to blame. However, when a blamer demonstrated bad commitment to a moral rule in their past behaviors, participants consistently granted the blamer low moral standing to blame. Low moral standing to blame was generally associated with perceiving the blame to be less effective and less likely to be accepted. Moreover, indirectly demonstrating moral commitment, such as acknowledging one's past wrongdoing and feeling/expressing guilt, modestly restored moral standing to blame. Our studies demonstrate moral commitment as a key mechanism for determining moral standing to blame and emphasize the importance of considering a blamer's moral standing as a crucial factor in fully understanding the psychology of blame.

The Effect of Information Exposure on Stigma Toward the COVID-19 Patient Mediated by Perceived Risk, Attribution of Blame and Protection Norm Conformity

This study employed the model of stigma communication (MSC) to analyze how exposure to COVID-19-related information affected stigma-related information sharing about people who contracted COVID-19 during the pandemic and examined the cognitive process of the MSC in a collectivist culture. Based on a survey of 526 social media users during the COVID-19 pandemic in China, the study found that exposure to contact tracing information and pandemic control information had different impact on stigma-related information sharing through a series of cognitive variables. A dual-path model showed that perceived personal risk influenced stigma-related information sharing through attribution of blame toward the infected (the personal path), while perceived social risk influenced stigma-related information sharing through protection norm conformity (the social path). Compared to the personal path, the social path is more salient in shaping stigmatized attitudes and behaviors. The findings and discussions added to our understanding of the intricate stigma communication process in a collectivist culture.

People's judgments of humans and robots in a classic moral dilemma

How do ordinary people evaluate robots that make morally significant decisions? Previous work has found both equal and different evaluations, and different ones in either direction. In 13 studies (N = 7670), we asked people to evaluate humans and robots that make decisions in norm conflicts (variants of the classic trolley dilemma). We examined several conditions that may influence whether moral evaluations of human and robot agents are the same or different: the type of moral judgment (norms vs. blame); the structure of the dilemma (side effect vs. means-end); salience of particular information (victim, outcome); culture (Japan vs. US); and encouraged empathy. Norms for humans and robots are broadly similar, but blame judgments show a robust asymmetry under one condition: Humans are blamed less than robots specifically for inaction decisions-here, refraining from sacrificing one person for the good of many. This asymmetry may emerge because people appreciate that the human faces an impossible decision and deserves mitigated blame for inaction; when evaluating a robot, such appreciation appears to be lacking. However, our evidence for this explanation is mixed. We discuss alternative explanations and offer methodological guidance for future work into people's moral judgment of robots and humans.

It's the AI's fault, not mine: Mind perception increases blame attribution to AI

Can artificial intelligences (AIs) be held accountable for moral transgressions? Current research examines how attributing human mind to AI influences the blame assignment to both the AI and the humans involved in real-world moral transgressions. We hypothesized that perceiving AI as having a human mind-like qualities would increase moral blame directed towards AI while decreasing blame attribution to human agents involved. Through three empirical studies-utilizing correlational methods with real-life inspired scenarios in Study 1 and employing experimental manipulations in Studies 2 and 3-our findings demonstrate that perceiving mind in AI increases the likelihood of blaming AIs for moral transgressions. We also explore whether it also diminishes the perceived culpability of human stakeholders, particularly the involved company. Our findings highlight the significance of AI mind perception as a key determinant in increasing blame attribution towards AI in instances of moral transgressions. Additionally, our research sheds light on the phenomenon of moral scapegoating, cautioning against the potential misuse of AI as a scapegoat for moral transgressions. These results emphasize the imperative of further investigating blame attribution assigned to AI entities.

Norm learning, teaching, and change

We present a broad notion of norms that can accommodate many of its interdisciplinary variants and offers a framework to ask questions about norm change. Rather than examining community norm change, we focus on changes in the individual's norm representations. These representations can be characterized by six properties (including as context specificity, deontic force, prevalence), and we examine which of the properties change as a result of norm learning and norm teaching. We first review research insights into norm learning based on observation, imitation, and various forms of inference. Then we examine norm learning that results from teaching, specifically teaching by modeling and demonstration, communication and instruction, and evaluative feedback. We finally speculate about how different kinds of norm change in a given community foster different kinds of norm learning in the individual community member.

Partisan Prejudice: The Role of Beliefs About the Unchanging Nature of Ideology and Partisans

Although there is a tendency to think all forms of essentialism-the belief that characteristics are inherent and unchangeable-are similar, some theories suggest different foundations and outcomes. We investigated if belief systems about the stability of political ideology (trait essentialism) and the fundamental nature of partisans (social essentialism) predict prejudice in opposite ways and if they do so via differential relations with blame. Across six studies (N = 2,231), we found that the more people believe the trait of political ideology is fixed (trait essentialism), the more they think that Republicans and Democrats are inherently different (social essentialism). Crucially, despite this positive correlation, trait essentialism was negatively linked to partisan prejudice and social essentialism was positively linked. The essentialism to prejudice links were driven, in part, by differential associations with blame attributions. Media messaging robustly influenced both types of essentialist thinking, with implications for prejudice.

"The way I see it makes me believe you intentionally did it": Intentionality ascription and gaze transition entropy in violent offenders

Cognitive processes underlying inferences regarding inferring mental states (i.e. intentionality ascription) are still to be investigated. To assess how people accumulate social cues in order to attribute intentionality, a measure of gaze transition entropy (GTE) seems indicated to throw some light on these processes. Violent behavior is associated with distorted attributional processes but also with deficiencies in attention to socially relevant cues. Therefore, the current study compared the level of entropy between both violent male and female offenders and non-offenders and explore the association between GTE and ascribing intentionality. The sample (N = 128) consisted of violent inmates (N = 63, 31 women) and adults living in the community (N = 65, 31 women). Lower entropy characterized violent offenders to a greater extent as compared to those with no history of volent crimes. Moreover, lower entropy predicted greater intentionality ascription especially in judging ambiguous and hostile harmful events but only in the violent offender group. Findings imply that hostile attributions in violent offenders not only depend on a predisposition to interpret external reality in a hostile manner but can be the result of an inferential processing based on insufficient and incomplete information.

The existence of manual mode increases human blame for AI mistakes

People are offloading many tasks to artificial intelligence (AI)-including driving, investing decisions, and medical choices-but it is human nature to want to maintain ultimate control. So even when using autonomous machines, people want a "manual mode", an option that shifts control back to themselves. Unfortunately, the mere existence of manual mode leads to more human blame when AI makes mistakes. When observers know that a human agent theoretically had the option to take control, the humans are assigned more responsibility, even when agents lack the time or ability to actually exert control, as with self-driving car crashes. Four experiments reveal that though people prefer having a manual mode, even if the AI mode is more efficient and adding the manual mode is more expensive (Study 1), the existence of a manual mode increases human blame (Studies 2a-3c). We examine two mediators for this effect: increased perceptions of causation and counterfactual cognition (Study 4). The results suggest that the human thirst for illusory control comes with real costs. Implications of AI decision-making are discussed.

Praise Is for Actions That Are Neither Expected nor Required

In six studies, we examined two foundational questions about moral praise. First, what makes an action praiseworthy? In Study 1, participants reported that actions that exceed duties (compared with dutiful actions) deserve greater praise and are perceived as less likely to happen. Second, what do observers infer from praise? Praise may communicate information about local norms. In Study 2, we found that-in general-participants expect praise to increase the likelihood of a behavior. However, in Studies 3-6, participants inferred that moral behavior that receives praise is less common and is less required and expected of people. These inferences led individuals to judge that someone would be less likely to perform a behavior that was praised. These studies provide insight into the lay beliefs and communicative function of moral praise.

Counterfactual simulation in causal cognition

How do people make causal judgments and assign responsibility? In this review article, I argue that counterfactual simulations are key. To simulate counterfactuals, we need three ingredients: a generative mental model of the world, the ability to perform interventions on that model, and the capacity to simulate the consequences of these interventions. The counterfactual simulation model (CSM) uses these ingredients to capture people's intuitive understanding of the physical and social world. In the physical domain, the CSM predicts people's causal judgments about dynamic collision events, complex situations that involve multiple causes, omissions as causes, and causes that sustain physical stability. In the social domain, the CSM predicts responsibility judgments in helping and hindering scenarios.

Body integrity dysphoria and moral responsibility: an interpretation of the scepticism regarding on-demand amputations

A patient who requests an amputation deemed medically unnecessary by professionals is disqualified per se from being regarded as having medical decision-making capacity. This decision is based on the assumption that there is an option to pursue something other than amputation; such an assumption in many cases overflows into therapeutic obstinacy. This is the case for individuals who have ill or damaged body parts and who wish to avoid recurrent and painful medical treatment designed to save the limb, as well as for individuals affected by body integrity dysphoria (BID). BID is a condition that is recognised by the WHO and is included in the International Classification of Diseases, 11th edition. Individuals who are affected develop an intense feeling of overcompleteness of their body configuration, which leads to the development of a strong sense of dysphoria and consequently the desire to amputate in order to remove the source of such discomfort. In the few cases in which amputation has been carried out, the results have proved successful; the individual's quality of life has improved and they have had no new amputation desires. No medical therapy, including medical amputation, is available currently for individuals affected by the condition. This situation leads many with BID to mutilate themselves. Such events create a challenging ethical dilemma for the medical world.The present paper is focused on the capacity of the individual with BID to do other than request amputation and the implications that this carries regarding moral responsibility. It is proposed that the autonomy of the patient cannot be disqualified by default based on the amputation request, despite its oddity, and that any scepticism demonstrated by the physicians is based on a false preconception of ill will or ignorance, which results in a blaming attitude towards the requesting person.

Observers' motivated sensitivity to stigmatized actors' intent

Does a harmful act appear more intentional-and worthy of opprobrium-if it was committed by a member of a stigmatized group? In two studies (N = 1,451), participants read scenarios in which an actor caused a homicide. We orthogonally manipulated the relative presence or absence of distal intent (a focus on the end) and proximal intent (a focus on the means) in the actor's mind. We also varied the actor's racial (Study 1) or political (Study 2) group. In both studies, participants judged the stigmatized actor more harshly than the non-stigmatized actor when the actor's level of intent was ambiguous (i.e., one form of intent was high and the other form of intent was low). These data suggest that observers apply a sliding threshold when judging an actor's intent and moral responsibility; whereas less-stigmatized actors elicit condemnation only when they cause the outcome with both types of intent in mind, more-stigmatized actors elicit condemnation when only one type, or even neither type (Study 2) of intent is in their mind. We discuss how these results enrich the literature on lay theories of intentionality.

Perpetrators' folk explanations of their regretted and justified aggressive behaviors

When people explain why they behaved aggressively, they can refer to their thought process that led to their aggressive behavior - so-called reason explanations - or to other factors that preceded their thought process - so-called causal history of reasons explanations. People's choice of what mode of explanation they give might be affected by whether they want to distance themselves (or not) from their past aggressive behaviors. To test these ideas, participants in the current study (N = 429) either recalled an aggressive behavior they regret or an aggressive behavior they believe was justified. Participants then explained why they behaved aggressively. Mostly, people gave reason explanations for their aggressive behaviors, which is consistent with past research on how people explain intentional behaviors. Further, and as predicted, participants who explained behaviors they believe were justified gave (relatively) more reason explanations and participants who explained behaviors they regretted gave (relatively) more causal history of reasons explanations. These findings are consistent with the idea that participants adjust their explanations to either provide a rationale for, or to distance themselves from, their past aggressive behaviors.

Understanding Blame in the Context of Childhood Obesity

This paper explicates blame as a psychological construct in communication processes, with a focus on its underlying structure and its mediating role between message characteristics and the public's issue engagement. Data were collected from Amazon Mechanical Turk (N = 373) via a Web-based experiment, where we manipulated a news story about a child suffering an obesity-related health risk (asthma attack vs. heart attack) with different levels of preventability (high vs. low) and severity (high vs. low). Findings showed that blame should best be conceptualized and operationalized as a latent construct comprising both cognitive and affective components. Blame mediated message effects on social responses related to obesity prevention, including punitive attitudes, policy support, and intentions toward interpersonal communication and civic participatory behaviors.

Probing the Causal Contribution of Reasoning to Third-Party Moral Judgment of Harm Transgressions

Recent work has supported the role of reasoning in third-party moral judgment of harm transgressions. In particular, reasoning may increase the weight of intention in moral judgment, especially following accidental harm, a situation that presumably requires judges to balance considerations about the outcome endured by a victim on the one hand and considerations about an agent's intention to cause harm on the other hand. Three preregistered lab-based studies aimed to test the causal contribution of reasoning to moral judgment of harm transgressions using experimental manipulations borrowed from the reasoning literature: time pressure (Experiment 1), cognitive load (Experiment 2), and priming (Experiment 3). Participants (N = 284) were presented with short fictitious scenarios in which the agent's intention toward a potential victim (harmful or neutral intent) and the action's outcome (victim's injury or no harm) were manipulated. Participants then reported their moral judgment of the agent's behavior (wrongness and deserved punishment) and their empathy toward the victim. We found that time pressure reduced judgment severity toward agents who had the intention to harm, but the reasoning manipulations overall did not impact judgment severity toward agents who harmed accidentally. We discuss why reasoning may sometimes influence how individuals account for intention in third-party moral judgment of harm transgressions.

People accept breaks in the causal chain between crime and punishment

Crime and punishment are usually connected. An agent intentionally causes harm, other people find out, and they punish the agent in response. We investigated whether people care about the integrity of this causal chain. Across seven experiments, participants (total N = 1,709) rated the acceptability of punishing agents for one crime when the agents had committed a different crime. Overall, participants generally approved of such wayward punishment. They endorsed it more strongly than punishing totally innocent agents, though they often approved of punishing agents for their correct crimes more strongly. Participants sometimes supported wayward punishment when wrongdoers were punished for a different kind of crime than the one committed, and they supported several different kinds of wayward punishments. Together the findings show that people often tolerate breaks in the causal chain between crime and punishment.

Highlighting Victim Vividness and External Attribution to Influence Policy Support Regarding the Opioid Epidemic: The Mediating Role of Emotions

Recognizing the need for more evidence-based interventions and the potential of well-crafted messages in communicating the opioid epidemic, this study investigates the effectiveness of two messaging strategies (i.e., victim vividness and external attribution) that have the potential to mitigate stigmatization and influence a wide range of public policies concerning the opioid epidemic. Building upon the attribution theory of interpersonal behavior, an experiment with a 2 (victim vividness: high vs. low) × 2 (external attribution: present vs. absent) between-subjects factorial design was conducted among a national sample of U.S. adults (N = 995). The findings show that the messages with greater victim vividness reduced support for victim-oriented punitive policies, whereas the messages that mentioned external attribution increased support for perpetrator-oriented punitive policies. In addition, the two messaging strategies also worked indirectly through various emotions to influence policy support. Discussions on this study's contributions to both theory and practice are provided.

People who seem disgusting seem more immoral

Despite unresolved questions about replicability, a substantial number of studies find that disgust influences and arises from evaluations of immoral behavior and people. Departing from prior emphases, the current research examines a novel, related question: Are people who are viewed as disgusting (i.e., people whose habits seem disgusting) perceived as more immoral than typical or unusual people? Four experiments examined this, also exploring the downstream impacts of moral character judgments. Adults who seemed disgusting were regarded as more immoral for purity and non-purity violations (Experiment 1) and less praiseworthy for prosocial acts (Experiment 2). In Experiment 3, an 8-year-old with typical (but seemingly disgusting) habits was rated as "naughtier" and likelier to misbehave than an atypical child who loved vegetables and disliked sweets. Experiment 4 revealed how, when no behavioral information is available, beliefs about target disgust influence beliefs about future behavior, helping explain why seemingly disgusting targets are viewed as more immoral, but not always more punishable for their bad behavior.

Distributing Blame Among Multiple Entities When Autonomous Technologies Cause Harm

As autonomous technology emerges, new variations in old questions arise. When autonomous technologies cause harm, who is to blame? The current studies compare reactions toward harms caused by human-controlled vehicles (HCVs) or human soldiers (HSs) to identical harms by autonomous vehicles (AVs) or autonomous robot soldiers. Drivers of HCVs, or HSs, were blamed more than mere users of AVs or HSs who outsourced their duties to ARSs. However, as human drivers/soldiers became less involved in (or were unaware of the preprogramming that led to) the harm, blame was redirected toward other entities (i.e., manufacturers and the tech company's executives), showing the opposite pattern as human drivers/soldiers. Results were robust to how blame was measured (i.e., degrees of blame versus apportionment of total blame). Overall, this research furthers the blame literature, raising questions about why, how (much), and to whom blame is assigned when multiple agents are potentially culpable.

Perpetrators', Victims', and Witnesses' Folk Explanations of Aggressive Behaviors

Unsurprisingly, victims and perpetrators often view aggressive behaviors differently. The current study examined whether victims, perpetrators, and witnesses also explained aggressive behaviors differently. The current study included 408 participants who recalled a time when they harmed another person (i.e., perpetrator memory), when another person harmed them (i.e., victim memory), and when they witnessed an aggressive behavior (i.e., witness memory). Replicating past research, participants rated their recalled aggressive behaviors from the victim perspective as being more harmful and less justified than they did for their recalled behaviors from the perpetrator perspective. When examining their explanations for the behaviors, participants most often explained their own aggressive behaviors by referring to their mental deliberations that led to their behavior (i.e., reason explanations). In comparison, they referred to background causal factors (i.e., causal history of reasons explanations)-such as personality traits, demographic factors, cultural norms, etc.-more when explaining others' aggressive behaviors, especially when the explanation was from the victim perspective. These findings show the subtleties in how people communicate about their aggressive interactions: When communicating about their own aggressive behaviors, people use modes of explanations that portray their behaviors as sensible, and when communicating about a time when another person behaved aggressively towards them, people use modes of explanations that omit the thought processes that led to those behaviors.

Prototypes of Victims of Workplace Harassment

What do people think of when they think of workplace harassment? In 13 pre-registered studies with French, British, and U.S. American adult participants (N = 3,892), we conducted a multi-method investigation into people's social prototypes of victims of workplace harassment. We found people imagined such victims in physically, socially, psychologically, and economically different ways compared with non-victims: for example, as less attractive, more introverted, and paid less. In addition, we found ambiguous harassment leveled against a prototypical (vs. non-prototypical) victim was more likely to be classified as harassment, and perceived to cause the victim more psychological pain. As such, both lay-people and professionals wanted to punish harassers of victims who "fit the prototype" more. Notably, providing people with instructions to ignore a victim's personal description and instead assess the harassment behavior did not reduce the prototype effect.

Blame framing and prior knowledge influence moral judgments for people involved in the Tulsa Race Massacre among a combined Oklahoma and UK sample

Introduction

How an event is framed impacts how people judge the morality of those involved, but prior knowledge can influence information processing about an event, which also can impact moral judgments. The current study explored how blame framing and self-reported prior knowledge of a historical act of racial violence, labeled as Riot, Massacre, or Event, impacted individual's cumulative moral judgments regarding the groups involved in the Tulsa Race Massacre (Black Tulsans, the Tulsa Police, and White Tulsans).

Methods and results

This study was collected in two cohorts including undergraduates attending the University of Oklahoma and individuals living in the United Kingdom. Participants were randomly assigned to a blame framing condition, read a factual summary of what happened in Tulsa in 1921, and then responded to various moral judgment items about each group. Individuals without prior knowledge had higher average Likert ratings (more blame) toward Black Tulsans and lower average Likert ratings (less blame) toward White Tulsans and the Tulsa Police compared to participants with prior knowledge. This finding was largest when what participants read was framed as a Massacre rather than a Riot or Event. We also found participants with prior knowledge significantly differed in how they made moral judgments across target groups; those with prior knowledge had lower average Likert ratings (less blame) for Black Tulsans and higher average Likert ratings (more blame) for White Tulsans on items pertaining to causal responsibility, intentionality, and punishment compared to participants without prior knowledge.

Discussion

Findings suggest that the effect of blame framing on moral judgments is dependent on prior knowledge. Implications for how people interpret both historical and new events involving harmful consequences are discussed.

The effect of wrongdoer's status on observer punishment recommendations: the mediating role of envy and the moderating role of belief in a just world

Our proposition postulates that the correlation between the wrongdoer's status and the punishment suggestions of onlookers is primarily influenced by group-oriented envy rather than the ascription of intentionality and is moderated by the belief in a just world. In three separate studies, 389 university students were asked to read scenarios describing a hit-and-run crime committed by either a rich or a poor individual and then report their opinions on intentionality attribution (Study 1 and Study 2), envy emotions (Study 2), punishment recommendations (all three studies), and belief in a just world (Study 3). Consistently, the findings indicated that those observing recommended harsher penalties to be imposed upon high-status perpetrators engaging in the same wrongdoing (such as hit-and-run) as their low-status equivalents. The effect of the rich receiving more severe punishment was predicted more strongly by envious emotions than by intentionality attributions to high-status wrongdoers and was only present for those observers who endorsed a lower belief in a just world.

Computational Social Psychology

Social psychologists attempt to explain how we interact by appealing to basic principles of how we think. To make good on this ambition, they are increasingly relying on an interconnected set of formal tools that model inference, attribution, value-guided decision making, and multi-agent interactions. By reviewing progress in each of these areas and highlighting the connections between them, we can better appreciate the structure of social thought and behavior, while also coming to understand when, why, and how formal tools can be useful for social psychologists.

Partisan animosity through the lens of blame: Partisan animosity can be reduced by a historicist thinking intervention

Partisan animosity has been on the rise in America. Partisan animosity involves blame, wherein political partisans blame outparty members for their beliefs and actions. Here, we examine whether a historicist thinking intervention-drawn from research on blame mitigation-can reduce partisan animosity. The intervention consisted of three components: (1) a narrative about the idiosyncratic development of one political opponent paired with (2) a message about how unique life experiences shape everyone's political beliefs and (3) a suggestion that outparty members can be changed by future formative experiences. Experiments 1 and 2 showed that the intervention reduced cold feelings-measured via Feeling Thermometer-towards the outparty for both Democrats and Republicans. Experiments 3 and 4 focused on more specific emotional changes. Experiment 3 showed that, for Democrats, the intervention increased compassion. Experiment 4 showed that, for Republicans, the intervention reduced disgust, disapproval, anger, and contempt, but had no impact on compassion. For Democrats, but not for Republicans, reductions in animosity were mediated by reduced perceptions of control of self-formation, the mediator identified in prior work on historicist thinking and blame mitigation.

Evaluative Deflation, Social Expectations, and the Zone of Moral Indifference

Acts that are considered undesirable standardly violate our expectations. In contrast, acts that count as morally desirable can either meet our expectations or exceed them. The zone in which an act can be morally desirable yet not exceed our expectations is what we call the zone of moral indifference, and it has so far been neglected. In this paper, we show that people can use positive terms in a deflated manner to refer to actions in the zone of moral indifference, whereas negative terms cannot be so interpreted.

If not me, then who? Responsibility and replacement

How do people hold others responsible? Responsibility judgments are affected not only by what actually happened, but also by what could have happened if things had turned out differently. Here, we look at how replaceability - the ease with which a person could have been replaced by someone else - affects responsibility. We develop the counterfactual replacement model, which runs simulations of alternative scenarios to determine the probability that the outcome would have differed if the person of interest had been replaced. The model predicts that a person is held more responsible, the more difficult it would have been to replace them. To test the model's predictions, we design a paradigm that quantitatively varies replaceability by manipulating the number of replacements and the probability with which each replacement would have been available. Across three experiments featuring increasingly complex scenarios, we show that the model explains participants' responsibility judgments well in both social and physical settings, and better than alternative models that rely only on features of what actually happened.

How people reason with counterfactual and causal explanations for Artificial Intelligence decisions in familiar and unfamiliar domains

Few empirical studies have examined how people understand counterfactual explanations for other people's decisions, for example, "if you had asked for a lower amount, your loan application would have been approved". Yet many current Artificial Intelligence (AI) decision support systems rely on counterfactual explanations to improve human understanding and trust. We compared counterfactual explanations to causal ones, i.e., "because you asked for a high amount, your loan application was not approved", for an AI's decisions in a familiar domain (alcohol and driving) and an unfamiliar one (chemical safety) in four experiments (n = 731). Participants were shown inputs to an AI system, its decisions, and an explanation for each decision; they attempted to predict the AI's decisions, or to make their own decisions. Participants judged counterfactual explanations more helpful than causal ones, but counterfactuals did not improve the accuracy of their predictions of the AI's decisions more than causals (Experiment 1). However, counterfactuals improved the accuracy of participants' own decisions more than causals (Experiment 2). When the AI's decisions were correct (Experiments 1 and 2), participants considered explanations more helpful and made more accurate judgements in the familiar domain than in the unfamiliar one; but when the AI's decisions were incorrect, they considered explanations less helpful and made fewer accurate judgements in the familiar domain than the unfamiliar one, whether they predicted the AI's decisions (Experiment 3a) or made their own decisions (Experiment 3b). The results corroborate the proposal that counterfactuals provide richer information than causals, because their mental representation includes more possibilities.

When guilt works: a comprehensive meta-analysis of guilt appeals

Introduction

Guilt appeals are widely used as a persuasive approach in various areas of practice. However, the strength and direction of the persuasive effects of guilt appeals are mixed, which could be influenced by theoretical and methodological factors.

Method

The present study is a comprehensive meta-analysis of 26 studies using a random-effects model to assess the persuasive effects of guilt appeals. In total, 127 effect sizes from seven types of persuasive outcomes (i.e., guilt, attitude, behavior, behavioral intention, non-guilt emotions, motivation, and cognition) were calculated based on 7,512 participants.

Results

The analysis showed a small effect size of guilt appeals [g = 0.19, 95% CI (0.10, 0.28)]. The effect of guilt appeals was moderated by the theoretical factors related to appraisal and coping of guilt arousal, including attributed responsibility, controllability and stability of the causal factors, the proximity of perceiver-victim relationship, recommendation of reparative behaviors, and different outcome types. The effect was also associated with methods used in different studies.

Discussion

Overall, the findings demonstrated the persuasive effects of guilt appeals, but theoretical and methodological factors should be considered in the design and testing of guilt appeals. We also discussed the practical implications of the findings.

Perpetrator Blame Attribution in Heterosexual Intimate Partner Violence: The Role of Gender and Perceived Injury

Partner violence resulting in physical injury is more often blamed on men than women for perpetrating the same offence, as men are often perceived to be more capable of inflicting injury. The current study used vignettes in a mixed-model design to examine the influence of perpetrator and observer gender, and weapon presence on observer blame. A split-plot analysis of variance produced a significant effect of perpetrator gender and an interaction effect of perpetrator gender and weapon presence. These findings suggest that perpetrator gender may be more important than weapon presence when examining observer perceptions.

Intelligence brings responsibility - Even smart AI assistants are held responsible

People will not hold cars responsible for traffic accidents, yet they do when artificial intelligence (AI) is involved. AI systems are held responsible when they act or merely advise a human agent. Does this mean that as soon as AI is involved responsibility follows? To find out, we examined whether purely instrumental AI systems stay clear of responsibility. We compared AI-powered with non-AI-powered car warning systems and measured their responsibility rating alongside their human users. Our findings show that responsibility is shared when the warning system is powered by AI but not by a purely mechanical system, even though people consider both systems as mere tools. Surprisingly, whether the warning prevents the accident introduces an outcome bias: the AI takes higher credit than blame depending on what the human manages or fails to do.

A Comparative Defense of Self-initiated Prospective Moral Answerability for Autonomous Robot harm

As artificial intelligence becomes more sophisticated and robots approach autonomous decision-making, debates about how to assign moral responsibility have gained importance, urgency, and sophistication. Answering Stenseke's (2022a) call for scaffolds that can help us classify views and commitments, we think the current debate space can be represented hierarchically, as answers to key questions. We use the resulting taxonomy of five stances to differentiate-and defend-what is known as the "blank check" proposal. According to this proposal, a person activating a robot could willingly make themselves answerable for whatever events ensue, even if those events stem from the robot's autonomous decision(s). This blank check solution was originally proposed in the context of automated warfare (Champagne & Tonkens, 2015), but we extend it to cover all robots. We argue that, because moral answerability in the blank check is accepted voluntarily and before bad outcomes are known, it proves superior to alternative ways of assigning blame. We end by highlighting how, in addition to being just, this self-initiated and prospective moral answerability for robot harm provides deterrence that the four other stances cannot match.

Jurors use mental state information to assess breach in negligence cases

To prove guilt, jurors in many countries must find that the criminal defendant acted with a particular mental state. However, this amateur form of mindreading is not supposed to occur in civil negligence trials. Instead, jurors should decide whether the defendant was negligent by looking only at his actions, and whether they were objectively reasonable under the circumstances. Even so, across four pre-registered studies (N = 782), we showed that mock jurors do not focus on actions alone. US mock jurors spontaneously rely on mental state information when evaluating negligence cases. In Study 1, jurors were given three negligence cases to judge, and were asked to evaluate whether a reasonably careful person would have foreseen the risk (foreseeability) and whether the defendant acted unreasonably (negligence). Across conditions, we also varied the extent and content of additional information about defendant's subjective mental state: jurors were provided with evidence that the defendant either thought the risk of a harm was high or was low, or were not provided with such information. Foreseeability and negligence scores increased when mock jurors were told the defendant thought there was a high risk, and negligence scores decreased when the defendant thought there was a low risk, compared to when no background mental state information was provided. In Study 2, we replicated these findings by using mild (as opposed to severe) harm cases. In Study 3, we tested an intervention aimed at reducing jurors' reliance on mental states, which consisted in raising jurors' awareness of potential hindsight bias in their evaluations. The intervention reduced mock juror reliance on mental states when assessing foreseeability when the defendant was described as knowing of a high risk, an effect replicated in Study 4. This research demonstrates that jurors rely on mental states to assess breach, regardless of what the legal doctrine says.

Gender biases in attributions of blame for workplace mistreatment: a video experiment on the effect of perpetrator and target gender

Introduction

Ambiguous psychological workplace mistreatment such as insulting or ignoring a co-worker might trigger gender bias. This study aims to examine whether female perpetrators receive more moral anger and blame from observers than men.

Methods

A sample of Austrian workforce members (n = 880, 55.00% women, 44.89% men, 0.11% diverse) responded to standardized videos showing a perpetrator's angry insult and a perpetrator's exclusion of a co-worker from lunch. In total, we edited 32 video clips with four female and four male professional actors. We manipulated the following variables: 2 perpetrator gender (male/female) * 2 target gender (male/female) * 2 types of mistreatment (insult/exclusion).

Results

As hypothesized, linear mixed-effects modeling revealed more moral anger and attributions of intent against female perpetrators than against men. Significant three-way interactions showed that female perpetrators were judged more harshly than men when the target was female and the mistreatment was exclusion. Female targets were blamed less when the perpetrator was female rather than male. Male targets did not evoke attributional biases. Observer gender had no significant interaction with perpetrator or target gender.

Discussion

Our findings suggest that gender biases in perpetrator-blaming are dependent on target gender and type of mistreatment. The stereotype of women having it out for other women or being "too sensitive" when mistreated by men requires more attention in organizational anti-bias trainings.

Human and machine drivers: Sharing control, sharing responsibility

Machines are empowered with ever-increasing agency and decision-making authority to augment or even replace humans in various settings, making responsibility attribution less straightforward when they cause harm. Focusing on their applications in transportation, we consider human judgments of responsibility for automated vehicle crashes through a cross-national survey (N = 1657) and design hypothetical crashes after the 2018 Uber automated vehicle crash reportedly caused by a distracted human driver and an inaccurate machine driver. We examine the association between automation level-the human and machine drivers have different levels of agency (i.e., the human as a supervisor, backup driver, and mere passenger, respectively)-and human responsibility through the lens of perceived human controllability. We show the negative association between automation level and human responsibility, partly mediated by perceived human controllability, regardless of the involved responsibility metric (rating and allocation), the nationality of the involved participant (China and South Korea), and crash severity (injury and fatality). When the human and machine drivers in a conditionally automated vehicle jointly cause a crash (e.g., the 2018 Uber crash), the human driver and car manufacturer are asked to share responsibility. Our findings imply that the driver-centric tort law needs to be control-centric. They offer insights for attributing human responsibility for crashes involving automated vehicles.

Outcome effects, moral luck and the hindsight bias

In a series of ten preregistered experiments (N = 2043), we investigate the effect of outcome valence on judgments of probability, negligence, and culpability - a phenomenon sometimes labelled moral (and legal) luck. We found that harmful outcomes, when contrasted with neutral outcomes, lead to an increased perceived probability of harm ex post, and consequently, to a greater attribution of negligence and culpability. Rather than simply postulating hindsight bias (as is common), we employ a variety of empirical means to demonstrate that the outcome-driven asymmetry across perceived probabilities constitutes a systematic cognitive distortion. We then explore three distinct strategies to alleviate the hindsight bias and its downstream effects on mens rea and culpability ascriptions. Not all strategies are successful, but some prove very promising. They should, we argue, be considered in criminal jurisprudence, where distortions due to the hindsight bias are likely considerable and deeply disconcerting.

Impact of partner phubbing on negative emotions: a daily diary study of mitigating factors

Interactions between romantic partners may be disturbed by a co-present mobile phone use when a partner ignores their interaction partner in favor of a smartphone. This common practice, called phubbing, promotes social rejection and exclusion, hence the partner who gets phubbed may report negative emotional experiences. However, these experiences may be buffered by a cognitive perception mechanism, when the partner’s behavior is still perceived as responsive (i.e., understanding or validating). Thus, we hypothesize that feeling understood or validated moderate the link between phubbing intensity and negative emotions. To test our hypotheses, we conducted a daily diary study over seven days, using a sample of N = 133 participants living with their partner. Multilevel modeling was applied, to examine between- and within-person processes. The findings indicate that perception of the partner as understanding and validating, despite the co-present mobile phone use, reduces the negative emotional experiences during phubbing, and the interaction effects indicate nuances between phubbing and understanding and validation by partner, which extend our theoretical comprehension and distinguish between the two as separate relationship-related constructs. Our research provides a unique insight into how mechanisms related to couple interactions may reduce negative experiences, a finding that may be useful in future interventions and couples' therapy.

Who attributes what to whom? Moral values and relational context shape causal attribution to the person or the situation

Eight preregistered studies (total N = 3,758) investigate the role of values and relational context in attributions for moral violations, focusing on the following questions: (1) Do people's values influence their attributions? (2) Do people's relationships with the violator (self, close other, distant other) influence their attributions? (3) Do the principles intrinsic to the violated values (e.g., loyalty to close others) further influence their attributions? We found that participants were more likely to attribute violations by distant others to the person committing the violation, rather than the situation in which the violation occurred, when participants endorsed the violated values themselves. The tendency to make dispositional attributions did not obtain for violations of participants' less highly endorsed moral values or non-moral values. Relationship with the violator also influenced participants' attributions-participants were more likely to attribute their own and close others' moral violations to situational factors, relative to distant others' violations. This relational pattern was pronounced for violations of "binding" moral values, in which protection of personal relationships and groups is primary. Collectively, these results support a relational-values account of causal attribution for moral violations, whereby attributions systematically vary based on (1) the relevance of the violated values to the attributor's moral values, (2) the attributor's personal relationship to the violator, and (3) an interaction between (1) and (2) such that the principles intrinsic to the violated values influence the effects of one's relationship to the violator.

Moral Judgments of Human vs. AI Agents in Moral Dilemmas

Artificial intelligence has quickly integrated into human society and its moral decision-making has also begun to slowly seep into our lives. The significance of moral judgment research on artificial intelligence behavior is becoming increasingly prominent. The present research aims at examining how people make moral judgments about the behavior of artificial intelligence agents in a trolley dilemma where people are usually driven by controlled cognitive processes, and in a footbridge dilemma where people are usually driven by automatic emotional responses. Through three experiments (n = 626), we found that in the trolley dilemma (Experiment 1), the agent type rather than the actual action influenced people's moral judgments. Specifically, participants rated AI agents' behavior as more immoral and deserving of more blame than humans' behavior. Conversely, in the footbridge dilemma (Experiment 2), the actual action rather than the agent type influenced people's moral judgments. Specifically, participants rated action (a utilitarian act) as less moral and permissible and more morally wrong and blameworthy than inaction (a deontological act). A mixed-design experiment provided a pattern of results consistent with Experiment 1 and Experiment 2 (Experiment 3). This suggests that in different types of moral dilemmas, people adapt different modes of moral judgment to artificial intelligence, this may be explained by that when people make moral judgments in different types of moral dilemmas, they are engaging different processing systems.

Explainable AI as evidence of fair decisions

This paper will propose that explanations are valuable to those impacted by a model's decisions (model patients) to the extent that they provide evidence that a past adverse decision was unfair. Under this proposal, we should favor models and explainability methods which generate counterfactuals of two types. The first type of counterfactual is positive evidence of fairness: a set of states under the control of the patient which (if changed) would have led to a beneficial decision. The second type of counterfactual is negative evidence of fairness: a set of irrelevant group or behavioral attributes which (if changed) would not have led to a beneficial decision. Each of these counterfactual statements is related to fairness, under the Liberal Egalitarian idea that treating one person differently than another is justified only on the basis of features which were plausibly under each person's control. Other aspects of an explanation, such as feature importance and actionable recourse, are not essential under this view, and need not be a goal of explainable AI.

Doesn't everybody jaywalk? On codified rules that are seldom followed and selectively punished

Rules are meant to apply equally to all within their jurisdiction. However, some rules are frequently broken without consequence for most. These rules are only occasionally enforced, often at the discretion of a third-party observer. We propose that these rules-whose violations are frequent, and enforcement is rare-constitute a unique subclass of explicitly codified rules, which we call 'phantom rules' (e.g., proscribing jaywalking). Their apparent punishability is ambiguous and particularly susceptible to third-party motives. Across six experiments, (N = 1440) we validated the existence of phantom rules and found evidence for their motivated enforcement. First, people played a modified Dictator Game with a novel frequently broken and rarely enforced rule (i.e., a phantom rule). People enforced this rule more often when the "dictator" was selfish (vs. fair) even though the rule only proscribed fractional offers (not selfishness). Then we turned to third person judgments of the U.S. legal system. We found these violations are recognizable to participants as both illegal and commonplace (Experiment 2), differentiable from violations of prototypical laws (Experiments 3) and enforced in a motivated way (Experiments 4a and 4b). Phantom rule violations (but not prototypical legal violations) are seen as more justifiably punished when the rule violator has also violated a social norm (vs. rule violation alone)-unless the motivation to punish has been satiated (Experiment 5). Phantom rules are frequently broken, codified rules. Consequently, their apparent punishability is ambiguous, and their enforcement is particularly susceptible to third party motives.

Relationship between attitudes toward COVID-19 infection, depression and anxiety: a cross-sectional survey in Japan

BACKGROUND

Although negative attitudes are known to develop with experiences of COVID-19 infection, it remains unclear whether such attitudes contribute to depression and anxiety as sequelae of COVID-19. We aimed to investigate the relationships between attitude towards COVID-19 infection and post-COVID-19 depression and anxiety.

METHODS

A cross-sectional survey of COVID-19 recovered patients was conducted from July to September 2021 in Japan. Outcome variables, depression and anxiety were assessed using the Patient Health Questionnaire-9 and Generalized Anxiety Disorder-7); scores of 10 and above were identified as having symptoms of depression and anxiety, respectively. Exposure variables were whether participants were experiencing the following attitude strongly: threat to life due to COVID-19 infection, helplessness regarding COVID-19 infection, blaming a third party who did not restrain from going outside, blaming themselves for their COVID-19 infection, worry about spreading the infection to others, and self-stigma (Self-Stigma Scale-Short). Modified Poisson regression analyses were performed to analyze the findings.

RESULTS

A total of 6016 responses were included in the analyses. The proportion of depression was 19.88%, and anxiety was 11.47%. The threat of life due to COVID-19 infection, helplessness regarding COVID-19 infection, blaming oneself for their COVID-19 infection, and self-stigma were significantly associated with depression and anxiety after adjusting covariates. Blaming the third party who did not restrain from going outside was associated with anxiety. There was no association between the worry about spreading infection to others and depression or anxiety.

CONCLUSION

Negative attitudes, including self-stigma with the experience of COVID-19 infection, were related to depression and anxiety. Further studies confirming whether countermeasures for preventing or decreasing the negative attitude towards COVID-19 infection mitigate these symptoms are needed.