Conference Schedule

What is the right thing to do? What is the right kind of person to be? Most philosophical reflection on morality centers on and presupposes a moral ideal, the ideal of a moral or virtuous person. Much empirical moral psychology also centers upon a unified account of human moral capacities, capacities that may be expressed differently within and across cultures.
This panel explores a different possibility: that individual communities may be made up of diverse “moral personalities” that may emerge in connection with one another forming some sort of “moral ecology.” If this diversity of personalities results in benefit to diverse individuals or to the larger community of which they are a part, it might be appropriately regarded as a “division of moral labor.”
While the idea of “complementarity” of moral dispositions is old (appearing in Plato’s Republic as well as in various “Ant and Grasshopper” parables) and is controversial (as in discussions of gendered “emotional labor”), until recently there has been little sustained moral psychological or philosophical work on this topic. This panel approaches the question and its implications drawing upon both empirical and philosophical considerations.

This study examined how moral values shape support for constrained (Just War) versus unrestricted (Dirty War) warfare in a Georgian sample (N = 634). Using Moral Foundations Theory, individualizing values (care, fairness) negatively predicted Dirty War support, while binding values (loyalty, authority, sanctity) positively predicted Just War support. Moral disengagement mediated these relationships, and war-related stress amplified harm-sensitive individuals' moral engagement. Findings suggest moral values influence how people support war, not merely whether they do.

Introduction. Misinformation is a global threat. It is widely seen as an epistemic problem, but our research shows it is also a moral problem. Despite widespread use of the word “misinformation” in political dialogue, little research has examined what lay people identify as misinformation—specifically, if people apply the term even to opinions they simply disagree with. People often interpret their moral beliefs as objective truths rather than personal views (Skitka et al., 2021). We propose that this tendency toward moral objectivity leads people to misperceive opposing moral opinions as factually incorrect, and therefore as misinformation that should be censored from political dialogue. Consequently, people’s well-intentioned concern about misinformation may reflect not only a concern about false facts, but also an intolerance for moral disagreement.

Objective. Across three pre-registered studies in the U.S. and an analysis of cross-cultural data, we show that moral objectivity predicts seeing misinformation as a threat and mislabeling opposing opinions as misinformation.

Overview of studies. Analysis of Pew Research Center data (N = 28,333) across 24 countries and six continents offered preliminary evidence for the association between moral judgment and misinformation perceptions. People who moralize controversial issues tend to perceive misinformation as a greater societal threat, a pattern that held on average across countries in a linear mixed-effects model accounting for country-level variation.

To examine the mechanism of this effect, Study 1 (N = 431) established the correlation between moral objectivity and misinformation perceptions. Across various controversial issues (e.g., abortion, religion), regression analyses showed people who perceived these issues as having objectively correct answers also perceived misinformation on them as more prevalent and problematic.
Study 2 (N = 242) showed that moral objectivity predicts seeing opposing moral opinions as misinformation. Participants read moral opinion statements that opposed their stance on abortion and climate change, then rated how much they saw each statement as objective, misinformation, and acceptable to censor. Linear mixed-effects regressions found that perceived objectivity was associated with more frequently mislabeling opposing moral opinions as misinformation, which then led to greater support for censoring these opinions. Seeing opposing moral opinions as harmful also mediated the relationship between objectivity and misinformation mislabeling.

Preliminary results from Study 3 (planned N = 400) provide support for the causal mechanism. Participants were randomly assigned to write about their views on abortion as either an objective fact or a subjective opinion. In a pilot sample (n = 173), we successfully manipulated the perceived objectivity of abortion views. Linear mixed-effects regressions suggested that participants in the objective condition perceived misinformation as a bigger threat and were more likely to label opposing opinions as misinformation compared to the subjective condition, which then predicted greater support for censoring those opinions.

Implications. This project is ongoing. Future studies plan to examine the role of harm perception in misinformation mislabeling and assess its prevalence in real-life political dialogue. Our findings suggest that solving the misinformation crisis may require not only correcting false facts but also addressing lay beliefs about moral objectivity.

References
Skitka, L. J., Hanson, B. E., Morgan, G. S., & Wisneski, D. C. (2021). The psychology of moral conviction. Annual Review of Psychology, 72, 347–366. https://doi.org/10.1146/annurev-psych-063020-030612

I would like to be considered for a data blitz or symposium talk.

We integrate and build on theories of ambiguity and intuitive moral judgment to explore third parties’ moral judgments toward leaders and their affiliated organizations following leader responses to misconduct allegations. Across one archival dataset (267,816 Tweets) and three experiments (N = 1,415), we compare the effect of ambiguous accounts (denials, silence) versus unambiguous accounts (apologies) provided by leaders accused of sexual misconduct. After confirming that denials and silence elicit more ambiguity than apologies (Study 1), we used machine learning techniques to examine moral sentiment expressed in Tweets toward accused leaders and found that those who denied the allegations or remained silent elicited more negative moral sentiment than those who provided an apology (Study 2). We replicated these findings in a follow-up experiment (Study 3). We also found that ambiguous accounts were indirectly more likely than unambiguous accounts to elicit third parties’ likelihood of corporate boycotting and publicly voicing concerns about the accused’s organization and avoiding future interactions with the accused via intuitive moral judgment (Study 4). In sum, this paper provides multi-method support for the intuition-based repercussions of ambiguous public accounts following misconduct allegations.

Trust judgments are central to moral cognition, and a long tradition of work has examined how shared identity shapes whom we trust. Two relevant literatures—on partisanship and on religious belief—have proceeded largely independently. Political psychology has documented that people trust co-partisans more than counter-partisans (affective polarization; Iyengar et al., 2019). Work on religion and cooperation has shown that people trust believers more than non-believers, even when they themselves do not believe, often interpreted through divine-accountability theory (Norenzayan, 2013).
We integrate these literatures by jointly varying a target's belief in God and party affiliation. We recruited 3,163 American partisans via Lucid, quota-matched to the national distribution on age, gender, ethnicity, and region. Participants rated trust in a hypothetical profile that varied independently in partisanship (Democrat vs. Republican), belief in God (believes vs. does not), and several irrelevant attributes (job, age, pet) added to disguise the study's focus. Trust was measured via three items (α=0.78) and analyzed with ANOVA including controls for participant and target demographics. The study was preregistered.
We found significant main effects of both profile partisanship (F=164.49, p<.0001) and profile belief in God (F=106.55, p<.0001), with the latter significantly moderated by participant partisanship and participant belief in God. Shared partisanship reliably increased trust across all subgroups (b=0.079 to 0.127; 19-30% increases; all p<.001). The effect of profile belief in God varied markedly. For Democrats who were not strong believers, belief in God had no effect on trust (b=0.001, p=0.95). For Democratic strong believers and Republican non-believers, belief in God mattered roughly as much as shared partisanship. For Republican strong believers, belief in God mattered more than shared partisanship (F=7.8, p=0.006)—so much so that they trusted Democratic believers more than Republican non-believers.
To probe mechanism, we examined how belief in God shaped perceptions of partisan authenticity. Participants broadly rated Republican believers as more "true Republicans" than Republican non-believers. A mediation analysis showed that perceived true-partisanship significantly mediated 26.9% of the interaction between participant partisanship and the effect of profile belief in God (indirect effect b=0.019, bootstrapped 95% CI [.012, .027]). This is consistent with belief in God serving partly as a signal of partisan commitment, particularly among Republicans.
These results carry implications for moral psychology. First, religious-cognitive signals can outweigh explicit partisan cues for substantial portions of the population, complicating accounts that treat partisanship as the dominant social identity. Second, the findings qualify the divine-accountability account: trust in believers is robust for some groups but absent for others, and is entangled with what belief signals about political coalitions. Third, they suggest a methodological blind spot. Partisanship has received vastly more empirical attention than religion as a basis for trust, yet for most subgroups belief in God is as or more important than party. The exception is low-belief Democrats—the demographic to which most academics belong (Duarte et al., 2014). To the extent that researchers' own intuitions shape which questions get asked, this may help explain the imbalance.

This paper examines the tension between Aristotelian virtue ethics, which views virtues as robust and consistent traits (Aristotle, 2000), and situationism, which emphasizes variability in moral behavior across contexts (Doris, 2002). We take up the challenge of making the situationist critique plausible in an evolutionary context. Drawing on Hamilton’s rule, game theory, and empirical evidence from anthropology and cognitive science, we show that spiteful behaviors become advantageous under conditions of scarcity, small population size, and local competition. Historical climate variability during the Pleistocene likely favored flexible behavioral responses over fixed moral dispositions, with heritable mechanisms involving dopamine and serotonin modulating social preferences. These findings undermine the assumption that cooperation is explained by stable virtues, supporting a situationist account of virtue-relevant behavior.

More specifically, we analyze four categories of social behavior—cooperation, altruism, selfishness, and spite—and emphasize the conditions under which each is favored (Hamilton, 1964). Our central focus is on spite, defined as behavior that imposes costs on both the actor and the recipient. Although seemingly irrational, spite can be evolutionarily advantageous when fitness is understood in relative terms, particularly under conditions of resource scarcity, small population size, and local competition (Gardner & West, 2004).

We point out that these conditions were common and variable throughout human evolutionary history, especially during the climatically unstable Pleistocene epoch (Gould, 1982; Timmermann et al., 2022). As a result, selection pressures would have favored individuals capable of flexibly shifting between cooperative and spiteful strategies depending on environmental conditions, rather than those with fixed behavioral dispositions. Anthropological evidence from hunter-gatherer societies supports this claim: cooperative norms tend to prevail during times of abundance but often break down during periods of scarcity, giving way to more self-interested or even destructive behaviors (Balikci, 1970; Boehm, 2012).

Further support comes from behavioral economics and psychology. Experimental games such as the ultimatum game show that individuals frequently engage in costly punishment driven by concerns for relative advantage, particularly when resources are scarce or competition is local (Henrich, 2000; Oosterbeek et al., 2004). These findings suggest that preferences for relative, rather than absolute, payoffs play a significant role in human social behavior (Eaton & Eswaran, 2003).

Finally, we explore potential neurobiological mechanisms underlying this flexibility. Evidence suggests that dopamine tracks relative reward value and is associated with increased selfish and spiteful behavior, while serotonin may signal resource availability and promote prosociality (Crockett et al., 2013; Pedroni et al., 2014). These systems, which have heritable components, may help explain how environmental conditions modulate social preferences and behavior.

We conclude that the best explanation of human social behavior is not the existence of robust virtues, but evolved sensitivity to situational variables. This challenges a core assumption of virtue ethics and supports a situationist account of moral psychology, according to which behavior is highly context-dependent and moral “virtue” is correspondingly fragile.

The advancement of research in moral psychology could be enhanced by greater engagement with psychological theories of goal-directed behavior and self-regulation (or goal-regulation for short). Theories of goal-regulation (Bandura 2001), including control theory and cybernetics (Carver & Scheier 1990), provide a framework for understanding how an organism (or system) regulates its actions according to a goal, where a goal is a representation of a “desired” state that “motivates” actions to achieve that state (i.e., discrepancy reduction). A goal in this sense is quite broad in scope: it can be short-term or long-term, can be a means to an end or an end-in-itself, can be a “bucket list” item or an enduring standard one lives by (e.g., being an honest person), etc. This goal-directed approach applies to diverse systems such as humans, animals, and machines – and this a strength of such theories. Of course, processes involved with human goal-regulation are quite complex, both in terms of how people strive for goals and importantly how people adopt new goals, or revise or reject existing goals (i.e., discrepancy production).

While processes of goal-regulation don’t necessarily provide a full explanation of behavior, they are likely part of any full explanation, and so that theoretical framework can provide insight into many different types of moral behavior. Furthermore, theories of goal-regulation can also provide some common theoretical tools to use across very different lines of research in moral psychology, and some advantages of this are in reducing “jingle” and “jangle” fallacies in research (Hanfstingl et al. 2024), and in helping to build a more cumulative approach to psychological science (Mischel 2008). In addition, since theories of goal-regulation apply so broadly (e.g., humans, animals, machines), they may also helpfully inform cross-cultural moral psychology, for while the content of goals and specific ways of striving for them surely vary across cultures, the more basic mechanisms of goal-regulation such as discrepancy reduction should remain the same.

As an example, there are numerous references in moral psychology to a “gap” between moral reasoning and action. The idea of the gap is that moral reasoning may be necessary, but not sufficient, to motivate moral action. This has prompted the search for other moral capacities that could bridge this gap and motivate people to act in accordance with their considered moral judgments (Darnell et al. 2019). However, the problem with framing this issue in terms of a gap between “moral” cognition and “moral” action is that it implies that acting on moral standards is significantly different from the rest of our goal-directed behavior. There’s likely nothing special about self-regulating according to moral goals that would give rise to a unique knowledge–action gap that wouldn’t also occur with non-moral goals. It’s already well-established that there is an intention-behavior gap more broadly (Webb & Sheeran 2006), and so a gap between moral judgment and action may often be a token instance of this more general phenomenon. This is just one of the ways in which theories of goal-regulation may benefit moral psychology research.

In this paper, I explore what makes for an adequate theory of well-being and then argue that reflection on the phenomenology of well-being delivers an adequate theory of well-being. I argue that an adequate theory of well-being must be enumerative, explanatory, and action-guiding. I then argue that distinctive to the experience of well-being is the mental state of resonance, from which we can develop an adequate theory of well-being.

Species-normal evolved baselines for development and functioning are used to discuss human behavior using Philippa Foot’s framing for assessing an organism: its essential nature, its development, the qualities that help the organism lead a full life, and the capacities that make it a proper member of its species. For the vast majority of our 300,000-year existence, humanity has lived communally nested (recently culturally derailed), with now empirically-verified positive effects on wellbeing and morality.
489/500 words:
To assess any organism, we must several aspects: its essential nature, its development, the qualities that help the organism lead a full life, and the kinds of capacities make it a proper member of its species (Foot, 2001; Thompson, 1995). Each species evolved to be socially and ecologically fitted, the former noted by Aristotle to be a prerequisite for virtue (Nussbaum, 1988).
Our species evolved from the social mammalian line, in existence for around 20-40 million years (Konner, 2005). Social mammals have particular needs that must be met to foster health and wellbeing. Humanity’s ancestral genes emerged from Africa and are carried by San Bushmen today, whose culture has been in existence for at least 150,000 years (Suzman, 2017).
The lifeway shared by egalitarian hunter-gatherers, like the Bushmen, all over the world has been described as the species-normal evolved nest (Narvaez, 2021; 2024; 2025). In babyhood and early childhood, intense nurturing is provided by a community of carers through welcoming support, nearly 24/7 touch and responsive care, and child-directed breastfeeding. But throughout life, community members live in welcoming responsive relationships, self-directed free play, partnership with Nature with regular healing ceremonies (Narvaez & Bradshaw, 2023). Holistic wellbeing is fostered through meeting basic needs from babyhood on and each practice is confirmed by the sciences today to be foundational to human health and flourishing (e.g., ibid).
Lifelong nestedness fosters species-normal character described by observers, anthropologists, and ethnographers (e.g., Ingold, 2005; Lewis, 2014; Sorenson, 1998; Widlok, 2017; Young, 2019). Those who have grown up in species-normal communal nestedness tend to have quiet minds, inner happiness and childlike awe. They are fully alive, honest, have a good sense of humor, and outstanding memories and sensory perception. They have a deep sense of community but also high autonomy. They can build habits at will, partner with their landscape to which they feel attached, and practice connection to the world of Spirit. With others, they tend to be highly
social, relationally attuned, enhancing the wellbeing of others; they demonstrate empathy and listen nonjudgmentally. They are helpful and generous. They offer unconditional love and forgiveness, show respect for ancestors, future generations and the web of life. And they are fiercely egalitarian. These qualities allow individuals to live cooperatively, with humans and non-humans, in various challenging ecologies, like the Kalahari Desert.
This compilation of species-normal characteristics from those who live in our ancestral context challenges contemporary ideas of moral character. Instead, we can see the lack of social fittedness increasingly common across the world with globalization of industrialized capitalism and the undermining of traditions that support human wellbeing. Relatedly, the lack of ecological fittedness, replaced with domination of Nature, is destroying planetary integrity, causing mass extinctions and universal toxic ecologies.
In our ancestral context of foraging bands, moral character as holistic virtue goes hand in hand with survival. Non-cooperation with Nature, whether human or non-human, would likely be a death knell at the individual and/or community level.

Large language models do not just answer queries, they make prior choices about how to frame an issue, and they prioritize what outputs to generate, making them implicit moral machines (Bonnefon et al. 2024). Sometimes they give outright racist answers, picking up on biases and stereotypes embedded in the data they are trained with (Peters 2025). Understanding how these models align with human moral values remains an open challenge. Prior studies probed LLMs through only Moral Foundations Theory (MFT) and standard sacrificial moral dilemmas. We evaluate fifteen state-of-the-art large language models by administering three value psychometric instruments - the Moral Foundations Questionnaire (MFQ) (Graham et al. 2009), the Morality-as-Cooperation (MAC) framework (Curry et al. 2019), and the Schwartz Value Survey (SVS) (Schwartz 1992) - to map each model’s underlying ethical profile. This is the most comprehensive set of values carried out on a large-scale, multi-instrument comparative analysis that spans both open-weight and proprietary models. Our key findings reveal that LLMs demonstrate high alignment with the values of Care, Universalism, Benevolence, Fairness, Reciprocity and Liberty, while showing significantly lower alignment with Loyalty, Authority, Power and Sanctity. This is consistent with research showing that WEIRD populations tend to prioritize individualizing and self-transcendent values (Care, Fairness, Universalism) over hierarchy and authority values (Graham et al., 2009; Henrich et al., 2010; Henrich 2020). These findings demonstrate the necessity of large-scale, multi-model evaluations for uncovering more culturally representative alignment strategies. Our results highlight the need for alignment strategies that account for pluralistic value sets and mitigate the propagation of normative monocultures in global AI systems. Although we do not expect human beings to hold multiple moral values, there is an open question whether we should require that LLMs do so, given their worldwide use (Haas et al. 2026).

Bonnefon, J. F., Rahwan, I., & Shariff, A. (2024). The moral psychology of artificial intelligence. Annual review of psychology, 75(1), 653-675.

Curry, O. S., Mullins, D. A., & Whitehouse, H. (2019). Is it good to cooperate? Testing the theory of morality-as-cooperation in 60 societies. Current anthropology, 60(1), 47-69.

Graham, Jesse, Jonathan Haidt, and Brian A. Nosek. 2009. “Liberals and Conservatives Rely on Different Sets of Moral Foundations.” Journal of Personality and Social Psychology 96 (5): 1029–46. https://doi.org/10.1037/a0015141.

Henrich, Joseph, Steven J. Heine, and Ara Norenzayan. “The Weirdest People in the World?” Behavioral and Brain Sciences 33, no. 2–3 (2010): 61–83. https://doi.org/10.1017/S0140525X0999152X.

Henrich, Joseph. 2020. The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous. New York: Farrar, Straus and Giroux.

Schwartz, Shalom H. 1992. “Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries.” In Advances in Experimental Social Psychology, 1–65. https://doi.org/10.1016/s0065-2601(08)60281-6.

Peters, Uwe, and Benjamin Chin-Yee. "Generalization Bias in Large Language Model Summarization of Scientific Research." ArXiv, (2025). Accessed May 18, 2026. https://arxiv.org/abs/2504.00025.

Large language models do not just answer queries, they make prior choices about how to frame an issue, and they prioritize what outputs to generate, making them implicit moral machines (Bonnefon et al. 2024). We evaluate fifteen state-of-the-art large language models by administering three value psychometric instruments - the Moral Foundations Questionnaire, the Morality-as-Cooperation framework, and the Schwartz Value Survey. Our key findings reveal that LLMs demonstrate high alignment with the values of Care, Universalism, Benevolence, Fairness, Reciprocity and Liberty, while showing significantly lower alignment with Loyalty, Authority, Power and Sanctity.

Humans constantly need to decide how to best allocate money, food, effort, rewards, and punishments. What is the fair way to divide a sum of money between two people in asymmetric positions? Our intuitions about how to fairly divide resources are often inconsistent: sometimes we think it is fair to reduce inequality through redistribution, favor the disadvantaged, and donate more to the needy. Other times we think it is fair to reward performance, effort, or ability, and favor those who bring more to the table. Past research highlights egalitarian and redistributive intuitions that favor the disadvantaged. Recently proposed contractualist theories propose that moral judgments mirror what rational agents would agree to, favoring parties in advantaged bargaining positions. We reconcile these views, showing people think it fair to favor the disadvantaged outside of bargaining contexts, but to favor the advantaged when they have more bargaining power: All else equal, those who enter negotiations in stronger positions, thanks to better alternatives, should get more. We thus identify a clear boundary explaining why our moral intuitions sometimes favor redistribution and other times exacerbate inequality. Specifically, across nine countries (the United States, Brazil, Ghana, Kenya, Nigeria, South Africa, Ukraine, India, and Pakistan) and 10 randomized experiments, participants (n = 6,304; 44,128 judgments) make third-party judgments about the morally best split of a fixed amount. We systematically vary the context (between-subjects: donation vs negotiation) and each party’s relative advantage (within-subjects: the price that each party obtains in the absence of agreement/donation). We compare the performance of seven cognitive models of moral judgment at both the aggregate and individual level. When the split is determined by a unilateral donation and existing inequalities do not alter bargaining positions, moral intuitions reflect redistributive concerns aimed at reducing inequality. When the split is negotiated (bilateral negotiation in Study 1; third-party negotiation in Study 2), moral judgments are completely reversed: they overwhelmingly track bargaining power and can be predicted with striking precision using classic rational bargaining models (e.g., the Nash bargaining solution, the Kalai-Smorodinsky solution). These findings demonstrate that fairness intuitions carefully track people’s underlying bargaining positions, favoring the advantaged when doing so is relevant to mutual benefit, but supporting the disadvantaged when bargaining is off the table. Even those who endorse egalitarian ideals may unintentionally reinforce “rich-get-richer” dynamics in negotiation settings. They also reconcile popular support for unequal market outcomes with simultaneous support for welfare and charity: both may result from an underlying moral psychology sensitive to bargaining power.

Several authors have argued for a cultural analysis of systems of moral judgment and decision-making ( Sachdeva, Singh, Medin, 2011). It has also been argued ( Shweder, Mahapatra & Miller, 1997) in a study conducted in India that different social positions have different types of moral responsibility. Despite some research, the examination of morality and its meaning and the role of class and culture have been rather scant. While the relationship between social class and morality has been extensively theorised in Western contexts, yet remains underexplored in societies characterised by rapid economic stratification, structural inequality, and cultural complexity, such as India.

In order to examine cultural nuances there is need to embed research in the indigenous traditions and knowledge systems of a country. In the Indian knowledge system, morality is understood through the concept of dharma, which refers to moral responsibility or duty towards self and society. As opposed to a fixed set of universal rules, 'dharma' is contextual and relational and is structured around to one's social position. Morality in this sense is not externally enforced but rather developed through self-awareness and inner discipline. Morality is beyond personal virtue, it is about acting in ways that sustain family, society, and the moral order of the world through one's role and situation. Particularly, extant research has often focused its inquiry at the two ends of the class spectrum, that is comparing the upper and the lower class while the middle class remains theoretically underexamined, despite being the fastest-growing and arguably the most psychologically complex class group in contemporary India. Through a series of three studies we examine how moral spaces are socially situated and contextually contingent. Study 1 showed that moral transgressions are differentially evaluated and responded to depending upon gender and age. Study 2 and 3 show how social class not only influences moral boundaries and spaces but also influence moral choices and evaluations. We also examine how social class inequalities impact lived experiences and definition of morality in the predominantly collectivistic culture in India. We present a model that maps the psychosocial mechanisms through which social class and culture shapes morality.

References

Sachdeva, S. , Singh,P. & Medin, D. (2011). International Journal of Psychology, 46 (3), 161–176.
Shweder, R. A., Mahapatra, M., & Miller, J. G. (1997). Culture and moral development. In J. Kagan (Ed.), The emergence of morality in young children. Chicago, IL: University of Chicago Press.

Despite substantial progress in moral psychology, the field remains constrained by a lack of diversity among participants, researchers, and theoretical frameworks. Historically reliant on Western, Educated, Industrialized, Rich, and Democratic (WEIRD) populations, the field faces notable challenges regarding the generalizability of its findings beyond a thin slice of human diversity. Here, we argue that advancing a truly global science of morality requires a tripartite approach to diversification: expanding participant populations to capture a broader spectrum of human values (going beyond WEIRD participants); diversifying the researcher base to incorporate varied cultural perspectives (i.e., going beyond WEIRD researchers); and broadening our cocneptual sampling, that is, studying moral phenomena beyond those tranditionally regarded as legitimate objects of investigation in the West. By adopting a stance of moral pluralism, researchers can shift from viewing cultural variation as a superficial aspect of morality to recognizing it as a vital source of insight. Moral pluralism goes beyond harm and individual rights (WEDIRD flavors of morality) to encompass a broader taxonomy of human morality that helps us understand conflicts, misunderstandings, and diverse cultural systems. Integrating varied perspectives enables a truly inclusive science of morality that acknowledges the diverse ways human societies organize moral life, moving the discipline beyond a WEIRD-biased template of right and wrong, and advancing and nurturing a truly international science of human morality.

Although labels’ connotations can exert a powerful influence on how labeled entities are evaluated and supported within the workplace (Ashforth and Humphrey, 1995, 1997), organizations and leaders do not have full control over these implicit meanings. Instead, labels’ connotations are shaped by the historical and cultural contexts in which they are used (Hall et al., 2015, 2021; Hall & Townsend, 2026). In particular, labels, like other words, acquire a positive or negative valence from the evaluative tone of the language with which they typically co-occur. This acquired valence is known in sociolinguistics as semantic prosody (Louw, 1993; Sinclair, 1991; Xiao & McEnery, 2006). Indeed, words can acquire negative (or positive) semantic prosody from the surrounding language, even if the words themselves are generally neutral in isolation (e.g., Hauser & Schwarz, 2016, 2018, 2023).
In the present research, we suggest that DEI and its component labels (diversity, equity, and inclusion) have divergent semantic prosodies due to their differing uses in public discourse. In particular, we posit that in the United States, DEI has emerged as the focal label used in discussions that frame efforts to increase diversity and create greater equity and inclusion in a deliberately negative light. For example, in this negative frame, DEI repeatedly co-occurs alongside words and phrases such as "illegal," "wasteful," or "radical agenda" (see Allen et al., 2025; Lawrence, 2024; The White House, 2025). In contrast, the independent labels of “diversity,” “equity,” or “inclusion” are used much less frequently in these discussions. As a result of this divergent use and co-occurrence, we propose that DEI has accumulated a more negative semantic prosody than its component labels, and, as a result, entities labeled with DEI will be perceived more negatively and supported less than entities labeled with its components, even when the underlying organization or initiative is identical.
We tested three predictions in this research across four studies – one archival study using natural language processing, one exploratory study, and two preregistered experiments. First, we predicted that DEI would be more negatively valenced (i.e., have a greater negative semantic prosody) compared to diversity, equity, or inclusion as indicated by the words with which these labels commonly co-occur (Study 1) and those with which they are associated (Study 2). Second, we predicted that DEI-labeled initiatives (Study 3) and organizations (Study 4) would receive less support than initiatives and organizations labeled with the component labels. That is, even when the content of the entity is identical, the greater negative semantic prosody of DEI (compared to diversity, equity, or inclusion) will result in lower support for the DEI-labeled entities (compared to the other-labeled entities). Third, we also predicted that these label-driven differences in support will be true among non-liberals (i.e., moderates and conservatives), but we expected no such difference among liberals (Studies 3 and 4).

Recent work on pragmatic encroachment (Fantl & McGrath 2009; Stanley 2005; Baril 2013) has extended the influence of practical considerations beyond knowledge and justification to epistemic excellence. This paper explores a further implication of pragmatic encroachment for one of the most important virtues associated with moral inquiry: open-mindedness.

Open-mindedness is typically understood as a disposition to seriously consider alternative viewpoints and to revise one’s beliefs when appropriate (Baehr 2011; Kwong 2016; Riggs 2010). While philosophers have devoted considerable attention to the nature, value, and limits of open-mindedness, relatively little attention has been paid to how practical stakes affect its proper exercise. Yet many of the contexts in which open-mindedness is regarded as especially important—moral disagreement, political conflict, public deliberation, and interpersonal judgment—are also contexts in which the costs of error vary dramatically.

I argue that if pragmatic encroachment is correct, then the demands of open-mindedness are themselves stake-sensitive. What open-mindedness requires of an agent depends not only on the evidence available to that agent but also on the practical significance of being wrong. In some circumstances, open-mindedness requires expanding inquiry, seeking additional perspectives, and suspending judgment. In others, it permits or even requires maintaining commitment and bringing inquiry to a close. The virtue is therefore not adequately understood as a fixed disposition toward greater consideration of alternatives. Rather, it is a context-sensitive norm governing inquiry under varying conditions of practical risk.

While Anne Baril (2013) has argued that practical considerations bear on whether a disposition counts as an epistemic excellence, I focus instead on the conditions governing the proper exercise of open-mindedness. Even holding fixed that open-mindedness is an epistemic virtue, what it requires of an agent varies with the practical stakes involved. This shifts the role of pragmatic encroachment from determining which traits count as virtues to governing how those virtues are properly exercised. More broadly, the paper illustrates how debates in epistemology can illuminate questions about moral judgment, deliberation, disagreement, and intellectual character. The result is a more fine-grained account of open-mindedness as a context-sensitive norm of inquiry, rather than a fixed disposition, thereby extending the scope of pragmatic encroachment within virtue epistemology and moral psychology.

Assuming that moral responsibility (in the desert-entailing sense) exists in our world, is it essentially historical? In other words, is it partially a function of how the agent came to be in the internal condition leading to a particular behavior? Midlife manipulation scenarios involving radical reversals of an agent’s values challenge “structuralist” accounts — accounts that answer this question negatively (e.g., Frankfurt 2002, Arpaly 2003, and Talbert 2009). In this paper, I build on this challenge and propose a new historical condition for responsibility which is both modest and comprehensive: Minimal Historicism. I start by establishing two desiderata for this kind of account. First, the condition should be complete, accounting for the intuitive absence (or presence) of responsibility in both radical reversals and less extreme manipulation cases without requiring that the agent had performed an action in order to “own” her values (which would invite regress objections). Second, the condition must be suitably connected to, and sufficient for, the exclusion of only responsibility-undermining circumstances.
I then analyze the debate between McKenna (2012, 2016) and Mele (2019), pointing to the limitations and the insights in their preferred versions of historicism. Mele’s negative historicism fails to meet the first desideratum, since it cannot account for less extreme yet responsibility-undermining manipulation, but it correctly emphasizes that a responsible agent cannot have had her values recently and suddenly reversed. McKenna’s positive historicism, on the other hand, mistakenly assumes that every bypassing of an agent’s mental states is responsibility-undermining, and for this reason it fails to meet the second desideratum. Despite this limitation, however, his account emphasizes the intuitive importance of opportunities to critically assess and sustain our values.
I argue that my Minimal Historicism can incorporate these insights and meet both desiderata. According to this proposal, an agent is responsible for a particular behavior only if (i) she had the opportunity to critically assess, endorse and sustain — or reject — the value-types whose instances play a role in producing that behavior, and (ii) the outcome of the opportunity (whether and how she takes advantage of it) corresponds in content to the role played by these instances in the behavior. Unlike stronger historical views, Minimal Historicism does not require psychological connectedness between temporal stages of the self. This becomes important in manipulation cases where an agent’s values are erased and “replaced” by others of the same type, yet the agent still seems morally responsible. These cases motivate a fundamental idea of Minimal Historicism: there must be something like a “functional connectedness” in a responsible person’s value-based agency. The paper clarifies this idea. My proposal also points to an interesting connection between the historical and the epistemic conditions for moral responsibility. Since values seem to include beliefs in their constitution, opportunities for critical assessment require at least the absence of epistemic bad luck and the agent’s minimal responsiveness to morally-relevant evidence.

Adding a normative component to philosophical theories of causation has recently gained popularity. I develop the empirical commitments these theories have regarding the norm effect on causation. The four main psychological hypotheses about the norm effect are at odds with these commitments. This could undermine these philosophical theories. I show, however, that they can account for the empirical data that motivates these hypotheses. I conclude by discussing consequences for the philosophical and empirical literatures.

Existing theories of morality suggest a varying number of moral values (e.g., Curry et al., 2016; Graham et al., 2013; Schein & Gray, 2012). However, when moral values come into conflict, there is no clear higher-order principle that dictates which moral value should be prioritized over others. Metamorality is a new theoretical framework that introduces principles for moral decision-making when moral values conflict. At least four metamoral principles have been suggested: Honor, Divinity, Utilitarianism, and Rational Self-interest (Atari & Zewail, 2025). In three pre-registered online studies, we developed and validated the Metamorality Scale and tested how metamoral principles are related to moral decision-making situations as well as cultural and cognitive factors. Studies 1 and 2 (total n = 536) tested how metamoralities are associated with moral foundations and moral decision-making in the U.S. In these studies, we demonstrated that each metamorality has a distinct pattern of associations with moral foundations and altruism, cooperation, parochialism, retaliation, and utilitarianism using hypothetical behavioral economic games. The results of ordinal and linear regression models showed that Divinity was associated with more altruism, Utilitarianism with more utilitarian decision-making, Honor with more deontological decision-making and more retaliation, and Self-interest with less altruism and cooperation. None of the metamoralities were significantly associated with parochial decision-making. Study 3 was a cross-cultural study with data from 13 countries (n = 3,674) where we explored cultural variation in metamoralities, and their associations with moral foundations, cultural and cognitive measures, and moral behaviors measured by incentivized economic games. We found that the Metamorality Scale had good internal reliability and validity across cultures. There was cross-cultural variation in the endorsement of the four metamoralities. While reliance on Honor, Divinity, and Self-interest increased with further cultural distance from the U.S., Utilitarianism was not significantly associated. Multilevel models revealed that Utilitarianism had robust positive associations with utilitarian and altruistic decision-making across cultures, while Self-interest had robust negative associations. Divinity and Honor displayed context-specific patterns for the two moral decision-making outcomes. These studies demonstrate how the theory of metamorality can advance our understanding of moral decision-making and illuminate the sources of moral conflict across the globe.

Children’s and Adults’ Reasoning About Antisocial Lying

It is well established that children and adults judge lying as wrong, yet little research has examined the reasons underlying these judgments. In the present study, children and adults evaluated and reasoned about two common forms of antisocial lying. Both samples referenced many different concerns in their reasoning, but children focused more on rules and punishment, whereas adults emphasized consequences for the self. Findings suggest reasoning about lying is heterogenous across development.

Abstract submission type: Poster presentation

Abstract

Background: Studies show that most school-aged children tell lies for self-benefit (i.e., antisocial lies), and that rates of lying remain high until adolescence (Carl & Bussey, 2019; Evans & Lee, 2011). It is also well-established that children believe antisocial lying is wrong to do (Talwar et al., 2002, 2016; Talwar & Lee, 2008). However, little research has examined the reasoning that underlies these judgments. Social domain theory (SDT; Turiel, 1983) research has shown individuals’ reasoning provides important information about the consequences and motivations they consider most important to their social judgments. Although people judge many different actions as wrong, their reasoning may emphasize different moral (i.e., concerns for others’ well-being), conventional (i.e. concerns for rules/punishment), and/or prudential features of a situation (i.e., concerns for consequences for the self). Furthermore, research has shown that with age, individuals may emphasize different types of concerns as their capacity for reasoning about abstract concepts develops (Nucci et al., 2017). Thus, identifying the reasons why children view lying as wrong, and whether these reasons shift from childhood to adulthood, may provide important information about the social-cognitive processes underlying dishonesty.

Method: The goal of the present study was to assess how children and adults reason about common forms of antisocial lying. We conducted semi-structured interviews with a sample of 4-8-year-old children (n=92; Mage=6.66; 50% female) in which they were presented two vignettes depicting concealment (e.g., lying to conceal making a mess) and instrumental lying (e.g., lying to win a prize). Following each story, children judged the lie’s acceptability (okay/not okay) and justified their judgment (reasoning, open-ended). A sample of undergraduate students (n=120; 82% female) evaluated identical vignettes through an online questionnaire and explained their reasoning using open textboxes. Reasoning responses were classified into several categories using a coding system informed by pilot testing and prior SDT research (Smetana et al., 2013). Superordinate categories included moral (harm, fairness), conventional (punishment, rules/authority), prudential (consequences for self), unelaborated (ambiguous/unclear), and other (responsibility, deservingness, interpersonal). Reasoning scores were computed for each category by averaging codes across the two stories.

Preliminary results: Results are presented in Table 1. Lying was judged as wrong by nearly all participants (95%). However, the types of concerns emphasized in their reasoning differed between children and adults. Children relied more on conventional reasons (e.g., concerns for punishment) than adults, whereas adults used more prudential reasons (e.g., concerns for learning/academics) than children. Several of the other-coded categories were used frequently and almost exclusively by adults, including concerns for personal responsibility, deservingness, and interpersonal consequences.

Significance: The preliminary findings suggest that adults, compared to children, are less concerned with punishment or rules prohibiting lying and place greater emphasis on a broader range of considerations. At the same, reasoning was heterogenous in both samples, indicating that lying is not viewed exclusively as a moral, conventional or prudential issue.

ABSTRACT (for data blitz and poster consideration) Moral hypocrisy typically entails inconsistent evaluations in which individuals deem their own transgressions as less objectionable while maintaining a harsher judgement of the same transgressions when enacted by others. In the present work, however, we explore the inverse of this asymmetry: instances in which individuals judge their own transgressive behavior more harshly relative to an identical transgression enacted by another person. Drawing on attributional accounts of moral judgment, the tendency for self-directed emotions to heighten self-scrutiny, and the possibility that condemning others is itself viewed as undesirable, we predicted that participants would judge self-enacted transgressions more severely than equivalent transgressions enacted by others. In a pre-registered between-subjects study (N = 422), we offer preliminary evidence for inverses in moral hypocrisy. The 22-item stimulus set included 18 reversal-eligible (RE) transgressions, which were developed from a qualitative pilot study in which undergraduate participants (N = 186) generated examples of behaviors they viewed as more acceptable for others than for themselves, and 4 universally condemnable (UC) transgressions, which were included as a comparison category and expected to elicit relatively stable evaluations across framing conditions. Both RE and UC items spanned the six moral foundations (Care/Harm, Fairness/Cheating, Loyalty/Betrayal, Sanctity/Degradation, Authority/Subversion, and Liberty/Oppression), allowing us to explore whether self-other asymmetries varied across domains. A linear mixed-effects model revealed that participants assigned to the self-framed condition (e.g., “Imagine that you decided to X...”) rated transgressive behaviors as significantly more immoral than participants assigned to the other-framed condition (e.g., “A person decided to X...”), Δ = 4.79, p < .001. Consistent with predictions, this asymmetry was observed for RE transgressions. Against expectations, however, a comparable pattern emerged for UC items despite the substantially greater severity of these transgressions. The overall interaction between framing condition and moral foundation neared significance (p = 0.08), with exploratory analyses suggesting that the asymmetry was attenuated for Fairness/Cheating items relative to other foundations. Overall, the present work grounds the phenomenon of inverses in moral hypocrisy and draws attention to the processes that may underlie increased self-scrutiny and/or increased tolerance toward the behaviors of others. Possible explanatory avenues are discussed and follow-up studies are outlined. (word count = 359)

Controversial politico-moral issues, like the Gaza conflict, are framed and debated through speech. It is easy to be compelled by people we agree with—but can we also be compelled by those we disagree with? One potential determinant of listener reactions is whether the speaker sounds morally reasonable (Lawlor, 2026; McMahon, 2009; Rawls, 1993; Samuelson & Dahl, 2026)—a careful, good-faith deliberator. Two studies examined (a) factors shaping perceptions of morally reasonable speech and (b) the social implications of those perceptions.

We expected that what makes people sound reasonable varies by agreement. When listeners agreed with the speaker, we more emotionally aroused speakers would be rated as more reasonable (H1), since they signaled commitment to a shared cause. But when participants disagreed with the speaker, speakers who acknowledged the issue’s complexity would be rated as more reasonable (H2).
In a pre-registered study, we presented US Adults with 16 real-world audio clips of people discussing the current Gaza conflict and asked about their general alignment too. The 16 clips formed a 2 (emotional intensity: high/low) × 2 (complexity: acknowledged/unacknowledged) × 2 (perspective: agree/disagree) within-subjects design. A validation study supported our intensity and nuance categorizations. Participants rated how reasonable and compelling each speaker seemed. We found support for H1, D(1) = 23.06, p < .001, with greater emotional intensity providing a greater reasonableness boost for agreed-with clips. H2 was also supported, Ds(1) = 464.58, p < .001, with nuance providing a greater reasonableness boost for disagreed-with clips. Reasonableness had social consequences: more reasonable speakers were rated as more compelling, D(1) = 3765.7, p < .001.

Emotional speech varies in quality as well as arousal (Dahl et al., 2014): An angry speaker may register as a threat, a sad speaker as a victim, and a calm speaker as someone ready to deliberate. Study 2 examined how these tones interact with agreement. We expected that disagreeing listeners would find calm or sad speakers more reasonable than angry ones (H3), and that for angry speakers, agreeing listeners who were angrier themselves would rate speakers as more reasonable (H4).

Using generative AI, we produced 144 audio clips in a 2 (stance: pro-Israel, pro-Palestine) × 2 (gender: male, female) × 3 (emotion: angry, calm, sad) design. Participants heard 12 clips and again indicated their general alignment on the conflict. A validation study confirmed adequate separation between tones. Another sample of US adults supported H3: among disagreed-with speakers, tone significantly influenced reasonableness, D(2) = 7.99, p = .018, with calm and sad rated above angry. Consistent with H4, angrier listeners rated angry agreed-with speakers as more reasonable, D(1) = 7.68, p < .001.Again, more reasonable speakers were more compelling, D(1) = 370.80, p < .001.

Our work shows that reasonableness depends on more than agreement: emotional tone of voice and acknowledgement of complexity also guides perceptions of moral reasonableness. Seeing someone as morally reasonable also implied seeing them as more compelling. These findings point to moral-reasonableness perceptions as an important, understudied feature of moral and political life.
 
References
Dahl, A., Sherlock, B. R., Campos, J. J., & Theunissen, F. E. (2014). Mothers’ tone of voice depends on the nature of infants’ transgressions. Emotion, 14(4), 651-665. https://doi.org/10.1037/a0036608
Lawlor, K. (2026). Being Reasonable: The Case for a Misunderstood Virtue. Harvard University Press.
McMahon, C. (2009). Reasonable disagreement: a theory of political morality. Cambridge University Press.
Rawls, J. (1993). Political liberalism. Columbia University Press.
Samuelson, A., & Dahl, A. (2026). Can reasonable people disagree about right and wrong? The determinants and implications of perceiving moral reasonableness. Cognition, 274, 106597. https://doi.org/https://doi.org/10.1016/j.cognition.2026.106597

We would like to be considered for a talk in a symposium. Thank you for your consideration.

In the United States, growing partisan sorting and the moralization of political identity have created conditions in which people may perceive diverse social groups as overlapping political coalitions. We propose the Moralized Coalitions Model (MCM), a framework describing how such perceptions shape social judgment. The MCM argues that when a salient identity, such as political party, is perceived as a coalitional hub, it becomes a cue for inferring membership in other groups (e.g., that a Democrat is also atheist or that a Republican is also a Southerner). When these perceived coalitions are moralized, evaluations of one group can spread to associated groups through perceived loyalty and affiliation rather than shared traits or ideology. Although the MCM is intended as a general framework, we test it in the contemporary U.S. partisan context, where social sorting and partisan moralization are especially pronounced.

Across five studies (total N = 2,226 U.S. adults and college students, including nationally quota-sampled and preregistered Prolific samples), we used behavioral tasks, surveys, qualitative interviews, and experiments to test the MCM.

Study 1 (N = 569) examined whether people view political party as informative about other identities. Participants treated party affiliation as more diagnostic than race, religion, gender, and other demographic categories. However, analyses of four national datasets (N = 28,908) showed that political party was no more predictive of demographic characteristics than several alternative identities. This gap suggests that people rely on coalitional inferences that exceed demographic realities.

Study 2 (N = 438) showed that participants reliably distinguished coalitional descriptors (e.g., wealthy) from trait-based descriptors (e.g., rational). Democrats and Republicans largely agreed about which social groups are associated with each party, but disagreed substantially about their corresponding traits.

Study 3 (N = 72) combined interviews and surveys and found that participants described parties more often by group memberships than traits, or policy issues. Participants who more strongly moralized politics were more likely to stereotype the outparty in these coalitional terms.

Study 4 (N = 348) tested whether endorsement of coalitional stereotypes predicts moral evaluations. Greater endorsement of these associations predicted more favorable moral evaluations of inparty-linked groups and less favorable evaluations of outparty-linked groups, even after controlling for baseline attitudes. For example, Democrat participants who perceived Christians as overlapping with Republicans evaluated Christians more negatively, whereas Republican participants evaluated them more positively.

Study 5 (N = 799) experimentally manipulated perceived overlap between parties and social groups. Increasing perceived overlap improved evaluations of groups associated with participants’ own party and worsened evaluations of groups associated with the opposing party. Mediation analyses indicated that these effects operated primarily through perceptions of loyalty, support, and affiliation rather than perceived ideological or trait similarity. Again, these effects were strongest among participants who moralized politics.

Together, these findings suggest that when social groups are perceived as part of moralized coalitions, evaluations extend beyond individual groups to broader networks of affiliated identities. The MCM offers a novel account of how perceived group structure and moralization jointly shape stereotyping, social evaluation, and intergroup conflict.

People are equipped with multiple strategies for moral decision-making, including moral rules and cost-benefit reasoning (Cushman, 2013). How does the mind determine which strategy to use, and how does that shape whose welfare people take into account? In the first part of my talk, I will present empirical evidence that moral decision-making is shaped by reinforcement learning operating over cognitive strategies (Maier, Cheung, & Lieder, 2025). Four preregistered experiments in which participants made a series of decisions in different realistic moral dilemmas pitting a moral rule against cost-benefit reasoning (total N = 2,328) showed that participants decreased their reliance on moral rules in favor of cost-benefit reasoning (or vice versa) depending on which of these two strategies produced morally better consequences in their view. This learning transferred to incentive-compatible donations (d = 0.17) and to convictions about utilitarianism versus deontology (d = 0.52). Computational modeling confirmed that these transfer effects were driven by many participants increasing their reliance on the moral strategy that had produced better outcomes in the previous decisions (metacognitive learning) rather than behavioral learning.
In the second part of my talk, I will argue that metacognitive reinforcement learning also adjusts the moral weight people assign to others' welfare. Our computer simulations (Spiteri, Maier, & Lieder, 2026) show that this learning mechanism can explain the findings of multiple cross-cultural studies according to which the introduction of markets expanded people’s moral circle from the extended family to strangers (Henrich et al., 2010; Baldassarri, 2020).
In the third part of my talk, I will show that this learning mechanism is gated by selective attention (Spiteri, Gao, Kim, Sun, & Lieder, under review). Across four preregistered experiments in which participants repeatedly allocated money between their ingroup and an outgroup and an internal meta-analysis pooling 11,603 participants, we found that directing attention to how a decision affected a distant outgroup increased subsequent generosity toward that group, whereas directing attention to ingroup consequences narrowed concern to the ingroup (intergroup contrast d = 0.56). These effects of attention on moral learning were partially mediated by increased empathic concern for the individuals affected by the highlighted consequences.
Finally, I will present ongoing research showing that structured reflection on the most significant consequences of a past decision can help people overcome these attentional biases. We devised a series of Socratic questions that guide people to reflect on their moral decision-making and how others were affected by the resulting decisions. Across five preregistered experiments (total N = 4,150), this guided reflection taught participants to use more deliberate and impartial strategies and increased their generosity toward charities helping distant others, animals, and the environment. These effects even transferred to some related entities outside the decision scenarios participants reflected on (d = 0.37) by expanding their moral circle. These effects also transferred to incentive-compatible donation decisions (d = 0.17). Taken together, these findings suggest that metacogntive moral learning offers a unifying account of how individuals and societies extend or retract moral consideration depending on which consequences they attend to and how they process them.

References

1. Baldassarri, D. (2020). Market integration accounts for local variation in generalized altruism in a nationwide lost-letter experiment. Proceedings of the National Academy of Sciences, 117(6), 2858-2863.
2. Cushman, F. (2013). Action, outcome, and value: A dual-system framework for morality. Personality and social psychology review, 17(3), 273-292.
3. Henrich, J., Ensminger, J., McElreath, R., Barr, A., Barrett, C., Bolyanatz, A., ... & Ziker, J. (2010). Markets, religion, community size, and the evolution of fairness and punishment. Science, 327(5972), 1480-1484.
4. Maier, M., Cheung, V., & Lieder, F. (2026). Learning from outcomes shapes reliance on moral rules versus cost–benefit reasoning. Nature Human Behaviour, 10(2), 268-287.
5. Spiteri, G. W., Maier, M., & Lieder, F. (2026). A computational theory of learning moral weights. CogSci 2026.
6. Spiteri, Gao, Kim, Sun, & Lieder (under review). Selective Attention Can Bias Prosocial Learning From Consequences.

This paper argues that we should be suspicious of assigning fundamental moral significance to causal connection as such. Drawing on recent work in causal cognition and counterfactual sampling, I argue that judgments about causal strength are shaped by factors like abnormality, salience, portability, and intervention-readiness. These features plausibly matter for prediction and control, but not obviously for fundamental morality. I argue that this gives us reason to distrust ordinary intuitions linking causal involvement and moral significance.

Causal Judgment and Counterfactual Reasoning of Rule Enforcement Across 11 Countries.

Presenter: Pui Yee Connie Chiu
Affiliation: The Graduate Center, CUNY
Co-Authors: Gantman, A; Wylie, J; McHugh, C; Choi, J; Park, H; Yaffe, N; Syropoulos, S; Paruzel-Czachura, M; Lukas, T; Amaya, S; Struchiner, N.
*We would like to be considered for a talk in a symposium

Brief Description
When rarely enforced rules were suddenly punished, people across 11 countries shifted blame from the rule-breaker to the enforcer. This shift was amplified in countries with less tolerance for norm deviation (i.e., tighter cultures). In cultures that viewed norms as more malleable than fixed, people were more likely to endorse counterfactuals suggesting that the punishment could have been avoided if the relationship between the enforcer and the rule-breaker had been different.

Abstract
When people break norms, they are judged as especially causally efficacious (Gill et al., 2022). As a result, when people are punished for a norm violation, they are more likely to be seen as the cause of that punishment. Yet many rules in society are frequently broken but rarely punished (e.g., jaywalking; Wylie & Gantman, 2023). In these cases, enforcing rather than breaking the rule may itself be counter-normative. Will people ever see the enforcer, rather than the rule-breaker, as the cause of the punishment? We recruited participants from 11 countries (N = 4,895) to test whether cultural variation in tolerance for breaking norms and rules affects causal attribution for punishment. We asked participants to evaluate a workplace vignette in which someone breaks a frequently violated but either rarely enforced or always enforced rule against taking pens from the office and incurs a fine (within-subject design). Data were analyzed using multilevel regression with random intercepts for participants and random intercepts and slopes for condition by country. Across countries, people were less likely to say that the rule-breaker caused the punishment when the rule was rarely enforced (b = -1.50, SE = 0.07, OR = 0.22, p < .001); however, the magnitude of this shift was stronger in cultures with greater sensitivity to norms (i.e., higher cultural tightness; b = - 0.17, SE = 0.08, p = .039). Cultures also varied in their endorsement of counterfactuals about avoiding punishment. Notably, in countries that perceived social norms to be more malleable than fixed, people were more likely to endorse counterfactuals suggesting the punishment could have been avoided if the relationship between the enforcer and the rule breaker was different (e.g., the punishment wouldn’t have happened if the enforcer had “looked the other way” or if they had had a better relationship; b = −0.33, SE = 0.11, p = .014). Together, these findings show that causal judgments and counterfactual reasoning about rules and punishment depend not only on rule violation itself but also on normative expectations about enforcement and cultural sensitivity to norm violations.

References
Gill, M., Kominsky, J. F., Icard, T. F., & Knobe, J. (2022). An interaction effect of norm violations on causal judgment. Cognition, 228, 105183.

Wylie, J., & Gantman, A. (2023). Doesn't everybody jaywalk? On codified rules that are seldom followed and selectively punished. Cognition, 231, 105323

In the fall of 2026, my _Concise Introduction to Moral Psychology_ will be published by Edward Elgar. This decidedly interdisciplinary book integrates conceptual distinctions and empirical evidence in order to build a comprehensive understanding of morality. It is a compact (200-page) textbook for advanced undergraduate and graduate students in psychology, cognitive science, philosophy, and sociology.

Moral psychology’s questions are as old as philosophy, and its methods and perspectives range from cognitive to social psychology, from developmental to neuroscience, from behavioral economics to computational modeling. So moral psychology is the multidisciplinary science of morality. But what is morality?

Scholars have debated whether there is one defining feature of the moral landscape, but the evidence suggests considerable diversity. This book introduces the diverse phenomena of this landscape and illustrates how moral psychology has investigated them.

Most scholars agree that morality is for group living. Successful groups endorse standards that guide individuals’ actions so they serve the interests of the group as a whole. We can call these collective standards normative systems, which include norms, values, rights, and nowadays institutions.

For normative systems to have impact, humans evolved individual-level capacities, including judgments, concepts, regulatory processes, emotions, and communication. Only with these capacities can people learn, follow, and enforce norms. Morality, then, is the union of group-level normative systems and individual-level capacities. This book surveys what we know about the foundational normative systems (ch. 1) and the five core moral capacities built on them: moral behavior (ch. 2), moral judgments (ch. 3), moral sanctions (ch. 4), moral emotions (ch. 5), and moral communication (ch. 6).

Chapter 1 analyzes what norms are and what functions they serve in human communities. We learn about the origins of norms in our biological relatives, in our distant and more recent human ancestors, and in human norm learners.

Chapter 2 examines what we know about the causes of moral and immoral behavior, deep in human evolution and in modern life. Encountering remarkable prosociality but also serial killers and genocide, we recognize that humans are the most moral and the most immoral creature we know.

Chapter 3 distinguishing between kinds of moral judgments, leading from norm judgments to fast and effortless moral evaluations, from wrongness judgments of action to blame judgments of persons. By acknowledging this diversity, we discover how affect, emotions, and judgments can meaningfully co-exist.

Chapter 4 traces the many options people have to regulate (praise, criticize, penalize) a person’s behavior, searching for the causal determinants of the mildest to the most severe tools of regulation.

Chapter 5 gathers the evidence for calling some emotions genuinely “moral.” Investigating which emotions are uniquely elicited by moral events and motivate moral behavior we learn how the capacities of moral judgment, behavior, emotion and regulation interrelate.

Chapter 6 reviews the vocabulary and functions of moral communication. It studies the rich communication about communication moral standards and departures thereof and the intricate sequences of blame management that build on criticism, apology, and forgiveness to repair social relationships.

*Brief Description*

In the fall of 2026, my _Concise Introduction to Moral Psychology_ will be published by Edward Elgar. This interdisciplinary book integrates conceptual distinctions and empirical evidence in order to build a comprehensive understanding of morality. It is a compact (200-page) textbook for advanced undergraduate and graduate students in psychology, philosophy, and cognitive science. The book dedicates one chapter each to morality’s foundation in norms and the five core moral capacities: moral behavior, judgments, sanctions, emotions, and communication.

Forgiveness is a valuable tool for overcoming negative emotions and healing relationships after
interpersonal conflict. We report the findings of two studies simultaneously conducted in the
United States and in Colombia with victims of war. Our findings support the emotional fading
account across cultures, showing that forgiveness is associated with decreased negativity during
retrieval while the episodic details are still remembered. Importantly, we found meaningful
cross-cultural differences, underscoring the importance of cross-cultural research on forgiveness.

Forgiveness is a valuable tool for overcoming negative emotions and healing relationships after
interpersonal conflict. The emotional fading account proposes that forgiveness does not involve
the forgetting of the episodic content of a remembered wrongdoing but, rather, the selective
modification of the emotional responses elicited at retrieval1. Recent evidence supports this
account by showing that remembering forgiven moral transgressions is experienced with
decreased affective responses compared to not forgiven ones, while the episodic details of the
memory are remembered just the same2. Unfortunately, this research has been confined not only
to WEIRD samples but also to wrongdoings that are relatively mild, limiting thus the
generalizability of the results. In this paper, we present the findings of two studies
simultaneously conducted with a student and community sample in the United States, and a rural
sample of victims of war in Colombia. Our results support the episodic fading account, and also
show important nuances between samples, underscoring the importance of cross-cultural
research.
In Study 1 (US: N = 110; Colombia: N = 255), we used a within-subjects design asking
participants to remember a neutral memory and a memory of a moral transgression. We
evaluated the phenomenological experience of remembering wrongdoings using in-depth
Autobiographical Interviews3 and its relationship with forgiveness using questionnaires to assess
interpersonal motivations after the transgression4. In study 2 (US: N = 140; Colombia: N = 88),
we used a similar within-subjects design comparing two neutral memories and two memories of
moral transgressions using both self-reported and psychophysiological measures. While
participants silently recalled each memory, we measured facial muscle responses of the
corrugator and the zygomaticus (markers of emotional negativity and positivity) using
electromyography, as well as skin conductance response (marker of emotional intensity).
Consistent with the emotional fading view of forgiveness, memories with high ratings of
forgiveness were associated with less negativity during recall in both the US and Colombian
samples across both studies. Importantly, forgiveness was not associated with differences in most
episodic details, suggesting that forgiveness reduces the negative emotional tone of memories
without altering their episodic characteristics (e.g., level of detail, coherence, location of
objects/people, imagery, relieving). These findings provide robust evidence for the cross-cultural
consistency of forgiveness as a psychological process that involves change in affective reactions
when remembering wrongdoings.
However, we also observed notable cross-cultural differences. Based on self-reported measures,
participants in the U.S. showed greater negativity during recall than participants in Colombia;
yet, automatic facial muscle responses, which serve as markers of experienced emotional
negativity, were significantly stronger in the Colombian sample. These findings are consistent
with positivity or presentation biases but, more importantly, they underline the importance of
using mixed methodologies to assess the negative emotional impact of episodes of wrongdoing.
Additionally, forgiveness was differently associated with the readiness to express negative
emotions. Whereas forgiveness was associated with less emotional expression in the U.S.
sample, Colombian participants showed more readiness to express their emotions. These
differences highlight the importance of approaching psychological research in the study of
forgiveness through a cross-cultural lens.

The target domain of morality should be identified according to the meaning of the folk concept of morality, not according to a definition based on theoretical principles. After explaining why theory-based definitions are inherently problematic, I illustrate how the folk concept can be used to identify the moral domain instead, borrowing methods from experimental philosophy. I present methods and results of an exploratory, cross-cultural study (N=1,613), along with a follow-up study (N=430).

Theory-of-mind development transforms how humans identify and understand conspecifics, yet its role in shaping perceptions of who is fully human remains understudied. We found among preschoolers, dehumanization emerged only in children who understood false beliefs and increased with mind attribution abilities. But among school-age children, advanced theory-of-mind performance was unrelated to either blatant or subtle dehumanization. These findings suggest that mind understanding is critical for the emergence of dehumanization, but not later variation in dehumanization.

Authorities often tell us what to believe. Sometimes their influence is epistemic: a teacher asserts a counterintuitive claim is true. Other times, authority operates through pressure rather than evidence: a parent threatens punishment for disagreement, or a political leader issues a moral directive. In both cases, individuals may comply; but does such compliance reflect genuine belief change? And should it?

These questions sit at the intersection of epistemology and moral psychology. Evidentialist philosophers argue that beliefs ought to be governed by evidence rather than pragmatic incentives (Clifford, 1877, Feldman and Conee, 1985), while involuntarists contend that beliefs cannot be brought about by direct act of will (Williams, 1970, Alston 1988). Together, these views imply that coercive influence over belief is both normatively illegitimate and practically ineffective. Yet laypeople often treat others' beliefs as controllable and subject to social pressure (Cusimano & Goodwin, 2020), and cultures differ both in how authority's legitimacy is construed (Ma et al., 2024) and in what minds are seen as free to do (Chernyak et al., 2013). Here, we explored whether children distinguish epistemic from coercive influence when reasoning about others' belief change, and whether culture shapes if such change is seen as possible or legitimate.

Six- to nine-year-old children in the United States (N = 98, Mage = 7.98) and China (N = 104, Mage = 8.05) heard vignettes in which the protagonist held an initial belief (e.g., water makes a plant grow), then encountered an authority figure who proposed an alternative (e.g., coffee makes a plant grow). The authority's influence was either epistemic (offering evidence) or coercive (threatening loss of a reward), manipulated between subjects. Across the four trials, the strength of the protagonist's prior belief also varied. For each trial, participants judged whether belief change was possible, whether the protagonist would change their belief, whether the protagonist would change their behavior to align with the new belief, and whether belief change was good or bad.

Generalized Linear Mixed- Effect Models revealed that children in both cultures distinguished epistemic from coercive influence: children judged belief change as less likely (p < .001)), predicted less behavioral change (p < .001), and deemed it less appropriate (p <.001) under coercion than under epistemic influence. Children from both cultures also recognized the constraints of prior belief strength: beliefs with weaker priors were rated as more possible to change, more likely to change, and more normatively appropriate to change than beliefs with stronger priors ( all ps <.001). Interestingly, children’s possibility judgments diverged between cultures. While US children judged belief change to be possible regardless of the type of influence (p = .323), Chinese children were generally less likely than US children to judge belief change as possible (p <.001), and specifically belief change to be less possible under coercion (p < .001). Together, these findings suggest that children's sensitivity to the distinction between epistemic and coercive grounds for belief develops early and is cross-culturally robust. Yet, what counts as doxastically possible—not merely impermissible—appears to be shaped by cultural experience early in childhood.

The nature of well-being is a central question in philosophical ethics, and lay conceptions of well-being are important for policy, which needs to respect the values of its beneficiaries. Following on earlier findings, we conducted two sets of studies to test lay support for three theories of well-being: hedonism, desire theories, and objective theories. In the first set, participants were presented with vignettes where they were asked to judge which of three options would be best for a given individual, either a 16-year-old or an adult (54), each corresponding to one of the three theories. A sample response option was “That Susan do what will result in a life in which she most realizes her potential, successfully pursuing worthwhile goals.”

Scenarios in one study focused on advice-giving, both for raising a child and for a career choice. The next study concerned retrospective assessment to confirm responses weren’t just focused on strategies for promoting well-being. Our findings suggest variation in the criteria for well-being people employ in different contexts, especially for different ages. Objective list theories and hedon-ism garnered the most support, for older and younger individuals respectively. Desire theories re-ceived shockingly little support.

These results already striking, but we will report on a further set of studies in progress that ask par-ticipants to assess multiple scenarios, each comparing how two individuals are doing in “mixed” vignettes where high well-being on one theory coexists in an individual with low well-being on another. Age is again varied, and each vignette is richly described to elicit judgments most similar to those made in ordinary life.

An example pairing good objective well-being with poor hedonic well-being: “Henry is a talented programmer who plays on a competitive soccer team. He has good relationships with his family and volunteers regularly at a nursing home. However, he has frequent headaches and often feels listless and empty. At night he often feels sad and has trouble sleeping.” The prompt asks partici-pants to rate whether Henry is doing better or worse than another person, who might also be high in objective well-being but is lower in desire-fulfillment.

For each theory of well-being three possible positive and three possible negative outcomes are ex-amined, so that each participant judges a subset of 18 possible combinations. The pattern of re-sponses across many such pairwise comparisons is expected to yield an unusually robust picture of lay support for each theory, as well as the effect of age on the well-being criteria deemed relevant for a given individual.

It is widely supposed that most people favor desire-fulfillment theories of well-being. Yet our find-ings contradict the received view, casting doubt on the acceptability of such theories to much of the public, despite their widespread use in policy settings. Our findings also offer suggestive evidence for contextualist or related conceptions of well-being among the folk, on which the ultimate stand-ards of well-being vary across contexts.

Moral psychology has long studied how intentions and outcomes shape blame, but overlooked a deeper question: how does moral cognition give rise to legal cognition? We introduce a novel paradigm to study this directly. In a cookie allocation task, children and adults face scenarios mapping onto tort law's three foundations—intentional wrongdoing, negligence, and strict liability—choosing to restore the victim, punish the perpetrator, or do nothing. This reveals when moral intuitions reflect legal categories.

Moral psychology's revitalization in the early 2000s was organized largely around utilitarian and deontological templates from post-Enlightenment moral philosophy, with dual-process models (Greene et al., 2001) framing moral cognition as a competition between affect-driven deontological intuitions and deliberative utilitarian reasoning. Yet growing evidence suggests this framework leaves central features of everyday moral judgment underspecified. People rarely apply explicit philosophical principles (Awad et al., 2020; Nichols & Mallon, 2006); instead, they use behavior—even seemingly harmless behavior—as a window into character (Royzman et al., 2015; Uhlmann & Zhu, 2014; Uhlmann et al., 2015). Even extraordinary altruists, such as people who donate kidneys to strangers, show no systematic utilitarian tendency (Amormino et al., 2022), complicating the implicit normative claim that consequentialist reasoning represents the morally superior mode.

We argue that Aristotle's virtue ethics offers a more empirically tractable framework for describing how people actually judge moral situations. Three Aristotelian commitments map onto current findings: (1) perception of particulars, which aligns with evidence that contextual features and relationships shape judgment (Hofmann et al., 2014; McManus et al., 2020); (2) habituated character, which helps explain why moral intuitions feel immediate even when verbal justifications are unavailable (Haidt, 2001); and (3) phronēsis, or practical wisdom, which reframes context-sensitivity from a philosophical slogan into a testable modeling target. We engage directly with the situationist objection (Doris, 2002), arguing that our proposal depends on the centrality of character inference in moral judgment rather than on a strong claim that people possess fully cross-situationally consistent traits—Miller's (2013, 2014) "mixed traits" account offers a productive middle path.

We then sketch a translational pathway for integrating virtue-ethical concepts into computational moral psychology. Virtues can be represented as latent dispositions—Bayesian priors or learned policies—that shape which features of a situation receive attention, how ambiguous evidence is interpreted, and which actions count as reasonable defaults. Reinforcement learning and Bayesian updating are natural computational homes for Aristotelian habituation: both describe how stable response tendencies emerge from repeated experience and generalize across novel contexts. Phronēsis becomes a computationally expressible competence—the capacity to map situational features onto context-sensitive action policies, rather than encoding a single global rule.

Following Hume's is–ought distinction, we do not claim that empirical evidence settles normative questions; rather, we treat virtue ethics as a historically rich source of testable hypotheses about the structure of moral cognition. Effective interdisciplinarity requires translation, shared vocabularies, and humility about disciplinary limits. Treated this way, virtue ethics can inform moral psychology and help bridge the humanities and sciences in the study of real-world moral judgment.

How does technology shape our understanding of human agency and, indeed, shape human agency itself? My focus is primarily neurotechnology, in particular brain-computer interfaces (BCI). Much of what I have to say has further bearing on agency in the context of predictive analytics and agentic artificial intelligence (AI). I argue that humans can do by thinking. Both conventionally via thought and trans-humanly as extended hybrid biological-artificial agent systems. Despite the arguably widely accepted and intuitive flavor of this conclusion, this talk argues that claiming this philosophical ground is surprisingly no easy matter. Understanding human agency in the context of BCIs and emerging smart technology requires confronting our cognitive dissonance about experimental interpretations and discarding traditional theoretical assumptions about cognition and agency. 

Brain-computer interface (BCI) technology promises to restore and transform our agency by decoding user intentions, even if the user lacks the capacity for intentional bodily movement. Users have been documented as experiencing a sense of agency, both explicit and implicit, over their BCI actions and outcomes, akin to the sense of agency experienced in the embodied case (Cornelio et al. 2022; Noel et al. 2023). In contrast, experimental decoding of intentions — such as the much-discussed broad Libet-inspired paradigm — has been taken to threaten agency, exposing the illusion of conscious control, even if the agent has a full range of intentional bodily movement. Intriguingly, both projects rely on the same neurotechnology and brain activity. Call the former Agency Optimism and the latter Perhaps Pessimism about BCI Agency. If these parallels between the BCI technology and Libet paradigms hold, then we ought to conclude the same thing metaphysically about BCI usage, even allowing for the practically transformative nature of the technology.

In this talk, I outline the best arguments for both the Agency Optimism and Perhaps Pessimism, including the Doing with Thought argument and Extended argument for Optimism and the Reliable Prediction Argument for Pessimism. I argue in favor of Agency Optimism about BCI Agency. To do so, I re-visit classic armchair thought experiments in action theory as well as neuroscience of agency. I outline four metaphysical candidate characterizations of (active) BCI use, as non-action events, as basic mental action, as non-basic action, and as basic extended action. The standard causalist story of action should cohere well with BCI action under the umbrella of basic mental action and nonbasic intentional action. However, I argue that on the standard causalist view, BCI users are only agents in the sense of consciously (or not) intending outcomes, a kind of mental action; their agency stops there, “in the head”. In response, I argue that the best metaphysical case for BCI use as agency involves adopting a hybrid sophisticated causalism plus extended mind approach, which I call Extended Causal Agency.

References
Cornelio, P., Haggard, P., Hornbaek, K., Georgiou, O., Bergström, J., Subramanian, S., & Obrist, M. (2022). The sense of agency in emerging technologies for human–computer integration: A review. Frontiers in Neuroscience, 16, 949138.

Noel, J. P., Bockbrader, M., Colachis, S., Solca, M., Orepic, P., Ganzer, P. D., ... & Serino, A. (2023). Human primary motor cortex indexes the onset of subjective intention in brain-machine-interface mediated actions. bioRxiv.

In her essays, 20th century Mexican philosopher Rosario Castellanos offers a compelling account of the ways in which patriarchal societies shape the conditions of women’s lives, not only by limiting their external opportunities, but by structuring the very possibilities through which they can understand and relate to themselves. On her view, women have historically been reduced to passivity through processes of mythification, confined to motherhood as the only avenue for agency, and burdened with an extreme and distorted ideal of self-sacrifice. Against this backdrop, Castellanos presents a conception of authenticity as an achievement: one that requires a woman to both reject the negation of false, imposed images of femininity and affirm her true, subjective personality.
I argue that Castellanos’s view gives rise to an agential dilemma: in the pursuit of authenticity, there is a risk that women will come to relate to themselves primarily through a rejection of conventional femininity, thereby reintroducing the very determinacy that Castellanos seeks to overcome. Additionally, a woman’s attempts to affirm her subjective personality risks collapsing into forms of negation or excluding aspects of herself that resemble the images she aims to reject.
The first horn of the dilemma concerns negation. In rejecting imposed images of femininity, women necessarily take those images as determinate objects of critique. Yet this very act of negation risks reintroducing determinacy at the level of self-conception: to reject a particular conception of femininity is, in part, to define oneself in opposition to it. So, one danger in Castellanos’s path to authenticity is that one can be trapped in the negation of the determinate and views oneself as existing outside of these negations. In doing so, one makes oneself determinate (which is just another image that is false in virtue of being determinate). To make oneself determinate is to put limits on oneself (e.g., I cannot be this thing because I am a woman, or because I am not a man, or because this thing is a part of conventional femininity that I find repulsive). If to be authentic is to be indeterminable, then creating oneself out of negation doesn’t seem to be a viable way to pursue authenticity.
The second horn concerns affirmation. Castellanos presents authenticity as requiring the discovery or affirmation of one’s ‘true personality’, but she also resists any strong notion of a fixed feminine essence or fully self-authored identity. Affirmation, therefore, cannot simply consist in the creation of a self ex nihilo, nor can it straightforwardly exclude elements of the self that resemble culturally imposed images. Yet without some criterion for distinguishing authentic affirmation from merely reactive self-definition, it becomes unclear how affirmation avoids collapsing into another form of negation or exclusion.
The resulting tension suggests that authenticity, as Castellanos conceives it, involves a delicate instability between negation and affirmation. I argue that this is not merely a feature of Castellanos’s account of femininity, but rather reveals a more general problem in the structure of authenticity itself.

This is a group symposium proposal with three papers. I've attached a pdf with:
--Symposium title: Evaluative and Moral Perception: New Philosophical and Empirical Perspectives
--Symposium speakers: (Mathew Fulkerson, Alon Hafri, Jonna Vance, and Preston Werner)
--The overall abstract (<500 words)
--Abstracts for each proposed paper (~75 words each)
--References

This paper examines whether anger can fittingly be directed at beneficiaries of historical injustice who are not themselves blameworthy for the original wrongdoing. I argue that anger need not always track culpability. In addition to accountability-focused anger, there exists a distinct form of “archival anger” that preserves moral memory and registers ongoing unjust relations. Because beneficiaries can occupy positions of enduring advantage created by past wrongdoing, they may be fitting objects of archival anger even in the absence of blame.

---

Anger is commonly understood as a response to culpable wrongdoing. Many influential accounts hold that anger is fitting because it tracks blameworthiness, holds wrongdoers accountable, or responds to violations of moral norms. Yet this view struggles to explain a familiar feature of moral life: people often experience anger toward beneficiaries of historical injustice even when those beneficiaries neither caused nor culpably perpetuate the original wrong. Can such anger be fitting?

This paper argues that beneficiaries of wrongdoing can be fitting objects of anger despite not being fitting targets of blame. I begin by examining contemporary defenses of anger, which emphasize its role in expressing commitment to moral values, preserving dignity, resisting injustice, and appreciating the significance of wrongdoing. While these accounts often focus on culpable agents, our ordinary moral practices suggest that anger is also directed toward institutions, social structures, and individuals who occupy positions created by unjust histories. Such cases indicate that not all fitting anger can be reduced to blame.

To explain this phenomenon, I introduce the concept of archival anger. Unlike paradigmatic forms of anger that respond to culpable actions and seek accountability, archival anger is directed at persisting unjust relations. Historical injustices frequently continue to shape sociopolitical and economic relations long after original perpetrators have died. Individuals may inherit advantages produced by these injustices without having caused or intentionally maintained them. Nevertheless, they can remain situated within ongoing relations of inequality. Archival anger responds to these enduring relations rather than to culpable acts.

I argue that archival anger serves a distinctive moral function. It preserves moral memory, acknowledges the continued presence of historical harms, and resists pressures to normalize or forget injustice. Its aim is not punishment, revenge, or blame attribution. Rather, it registers that a moral wrong remains socially significant and insufficiently resolved. Because beneficiaries can occupy positions constituted by enduring unjust relations, they can become fitting objects of archival anger even when they are not blameworthy.

The paper concludes that anger’s fittingness conditions are broader than commonly assumed. While some forms of anger appropriately track culpability, others track persistent moral realities. Recognizing archival anger helps explain why anger toward beneficiaries of historical injustice need not be irrational or merely resentful. Instead, it can constitute an appropriate moral response to enduring structures of injustice.

This study examined how Confucian virtue beliefs influence lay judge decision-making. Utilizing a simulated criminal case with 295 Taiwanese participants, researchers found that stronger Confucian beliefs predicted harsher sentencing for defendants with past negative behavior or familial ties to the victim. These findings reveal that cultural values profoundly shape judicial judgments, demonstrating that Confucian lay judges heavily weigh both a defendant's moral character and their interpersonal role obligations when determining guilt.

Despite extensive research on moral disagreement, we lack a clear empirical picture of what these conflicts look like as they unfold in ordinary social life. Across three studies (N = 734), participants described their most recent disagreement along dimensions of context, content, and consequence. Disagreements most often arose in close relationships and personal settings, and people were surprisingly tolerant and open to updating their views. Resolution approaches predicted how the disagreement felt but not whether minds changed.

As wartime decisions and targeting increasingly fall to autonomous weapons systems (AWS), this raises a question of public response to AWS versus human military action. The ethical issues of AWS and their conflict with international humanitarian law (IHL) are well documented (Sharkey 2012; Muller 2016; Asaro 2020). In moral psychology, testing on autonomous decision-making causing harm has examined whether autonomous weapons create a responsibility gap, alongside evidence of a general aversion to machine decision-making attributed to a lack of perceived responsibility and intentionality, with mixed findings (Bigman and Gray 2018; Robbins 2025). Public attitude surveys further find that support for or opposition to AWS is highly contextual, based on perceived military utility or other benefits (Horowitz 2016; Young and Carpenter 2018; Ada Lovelace Institute and Alan Turing Institute 2023).

However, AI-committed war crimes, particularly compared to human-committed ones, remain underexplored empirically, and prior work has been predominantly perception-based rather than behaviorally tested. In the next war, will the public call for a different political or military response to a war crime committed by an AI or a human? This study compares public retaliation and escalation preferences toward a military action violating IHL depending on the decision-maker, using 'talosian' and 'anthropogenic' as a descriptive distinction for a war crime committed by an AI versus a human perpetrator.

This study uses "Just War," a custom serious game in which participants lead a country at war against an unnamed fictional country, in an interactive branching-choice text narrative (Majnemer and Meibauer 2023). Participants (n = 297, adults recruited through Prolific) play eight scenarios with narrative consequentialism, where choices affect the ending, in a within-subject 4×2 design (war crime: attack on aid, non-combatant targeting, torture, chemical and biological warfare × perpetrator: human vs. AI). Each scenario is rooted in a specific IHL violation and real-world technology development. Participants respond with actions coded to a custom escalation ladder from inaction to retaliatory war crime, used as a behavioral index of moral acceptability of retaliation (Kahn and Schelling 2009). This gamification is a design choice to increase emotional and moral engagement, as opposed to standard moral psychology and political science vignettes (FeldmanHall et al. 2012; Triantoro et al. 2020; Lyreskog et al. 2023).

A repeated-measures ANOVA revealed an interaction between perpetrator and war crime type: retaliation against an AI was not uniformly higher or lower than against a human, but varied by violation. For the non-combatant targeting and chemical and biological warfare pairs, participants had higher escalatory responses against the talosian perpetrator. For the torture and attacks on aid pairs, participants had higher escalatory responses against the anthropogenic perpetrator. The study offers two contributions: a comparison of retaliatory preferences toward talosian versus anthropogenic war crimes, rather than the standard isolated consideration of AI military action; and a novel methodological design using narrative-consequentialism serious-game research. As wartime targeting is increasingly placed in algorithms instead of humans, researching moral judgment and retaliation preferences is critical to understanding how the next technology-enabled conflict may develop.

Loneliness is often described as a painful response to the absence of social connection. This paper argues that a central form of loneliness is better understood as a fitting negative emotion that arises when one reasonably takes oneself to lack attitudinal recognition of one’s relational irreplaceability by at least one qualified other.
The paper focuses on relational loneliness, understood as the loneliness that arises from lacking a desirable form of togetherness within intimate relationships. Existing recognition-based accounts offer the most promising framework for understanding this phenomenon. One view, inspired by Arendt and developed in different ways by Velleman and Setiya, explains loneliness as a failure to have one’s dignity or worth as a person recognized. A second view, defended by Creasy and Yao, argues that loneliness is not primarily about the nonrecognition of generic human worth, but about the failure to be recognized in one’s particularity: as the concrete individual one is.
I argue that both approaches capture something important but remain incomplete. Dignity-recognition views are too impersonal to explain why one may be treated with full respect and still feel lonely. Lack of respect may generate humiliation, indignation, or a sense of social invisibility, but these are not identical to loneliness. Particularity-recognition views come closer to the target, but they risk making loneliness too cognitive. One may be accurately understood, sympathetically interpreted, and even appreciated for one’s distinctive traits, while still feeling lonely if the other’s stance remains detached. What is missing is not merely correct appraisal, but a caring orientation in which one matters to another in a non-substitutable way.
I develop this thought through three notions. First, relational irreplaceability is not numerical uniqueness or functional indispensability. It is the status one has when, within a caring relationship, one occupies a non-substitutable place in another person’s practical and emotional life. Second, attitudinal recognition is not a detached judgment that someone has value, but a caring stance constituted by concern, responsiveness, and emotional attachment. Third, the relevant recognition must come from a qualified other, who is distinct from oneself, responsive to another’s value, capable of love and care, and situated in a special relationship through which her attitudes can recognize one’s relational standing.
This account has several implications for moral psychology. It explains why instrumental relationships often intensify loneliness, since being valued merely for one’s usefulness leaves one open to substitution. It also clarifies why loneliness reveals both human vulnerability and limits on individual agency. Since the relevant good depends on another’s freely given care, there can be social responsibilities to secure opportunities for relationship formation, but no straightforward right not to be lonely. Finally, the account helps explain why AI companions and social robots may alleviate the feeling of loneliness without resolving relational loneliness: they may simulate responsiveness without providing genuine attitudinal recognition of relational irreplaceability.

Research in moral psychology has focused on the living, examining how victims’ suffering (Gray, Schein & Ward, 2014), causing harm (Greene, 2014), and social relationships (Curry, 2016) drive moral judgments. The dead have received little attention, appearing primarily in research on disgust-based moral judgment and pathogen-avoidance (Haidt et al., 1993; Inbar & Pizarro, 2022).
However, legal systems and everyday practices often treat the wishes of the dead as binding–even when this goes against the interests of the living (Smolensky, 2009). Philosophers have debated whether this can be justified within a secular worldview (Lam, 2019). Proponents of these moral duties have typically explained them in terms of posthumous welfare (Feinberg, 1980, 1987; Pitcher, 1984; Luper, 2004, 2018).

In a series of preregistered between-subjects experiments (total N = 1663), this paper examines the underexplored moral psychology of respect for the dead.

Participants were recruited on Prolific and randomly assigned to minimally manipulated text-based vignettes. Attention and comprehension checks ensured data quality. Materials, data and code are shared on OSF.

Study 1 (N = 538) data were analysed using linear mixed models and showed that participants see violating a person’s wishes as highly wrong irrespective of whether that person is alive or dead (and irrespective of whether the wish pertains to property, money, privacy, or bodily integrity), with wrongness perceptions only slightly reduced if they died 100 years ago.

Study 2 (N = 580) examined whether perceived moral obligations to fulfil a wish were robust to convenience-based or moral costs. While agents are perceived to have weaker duties to fulfil severely immoral wishes, even severe inconvenience did little to mitigate the perceived moral obligation to fulfil the wishes of the dead.

Study 3 (N = 545, UK representative for Sex, Age, Ethnicity) found that going against wishes is viewed as disrespectful and disloyal, but not as impacting the welfare of the dead. Moreover, welfare and harm-related considerations did not explain moral judgments. Only 2% of participants explained their moral concerns in terms of harm rather than respect; perceived disloyalty and disrespect, were predictive of moral judgments in multiple regressions, but not harm; loyalty was the only MFQ foundation that was predictive (but explained little variance; Graham et al., 2008, 2011; Iyer et al., 2012); and neither dualism (Stanovich, 1989) nor post-mortem mind perception (Doyle & Gray, 2020) explained moral judgments.

Overall, participants judged that others' wishes generate strong moral obligations regardless of whether the person is alive or dead, the type of wish involved, or the costs of compliance. Further, contrary to much work in moral philosophy, these moral judgments were not related to any perception that the dead could be benefited or harmed, nor were these judgments explained by religious beliefs, politics, mind-body dualism, or mind perception. Instead, duties to fulfil others’ wishes were viewed as supported by the values of respect and loyalty – both when someone is alive and on into death. These findings suggest that harmlessly wronging the dead comprises an important and understudied domain of secular moral psychology.

As human beings, we not only live our lives but also reflect on life as a whole. Yet moral
psychology has traditionally focused on judgments of actions and behaviors in specific contexts.
What gives life meaning? What is happiness? What does it mean to live a good life? These
questions are often treated as matters of personal preference whose answers vary
idiosyncratically across individuals. For example, some people may cite their family as the
source of meaning and happiness in their lives, whereas others may point to their hobbies, work,
or personal passions. In this talk, I propose a different view. Although people may each have
their own answers to life's big questions, their judgments are not arbitrary. Instead, they are
guided by shared evaluative concerns, among which moral judgments play a central role.
Across research on happiness, meaning, and moral obligations, we find that moral considerations
systematically shape how people engage with and evaluate answers to life's big questions such as
happiness and meaning. Research on happiness suggests that people do not view happiness as
mere pleasure or desire satisfaction. Instead, both children and adults take moral character into
account when judging who is happy, often judging bad people to be less happy even when those
individuals obtain everything they want (Phillips et al., 2011, 2017; Yang et al., 2023; Yang,
2025). Research on meaning further reveals that people place even greater emphasis on selftranscendence
and contribution to society when making judgments about meaning, suggesting
that lives and activities are viewed as more meaningful when they benefit others or extend
beyond the self (Huang & Yang, 2023; Ge & Yang, 2023). Together, these findings suggest that
morality plays an important role in how people understand happiness and meaning in life.
Yet these findings raise a further question: What kind of morality shapes these evaluations?
Traditional moral psychology has largely focused on how people treat others. However, our
research suggests that both children and adults recognize obligations to the self (Soter et al.,
2025). Moreover, evaluations of happiness (Chen et al., 2023), meaning (Huang & Yang, 2023),
and life pursuits (Kim & Yang, 2025) are shaped not only by how people treat others but also by
how they treat themselves. Recognizing this broader conception of morality, one that includes
obligations to the self, thus helps illuminate how people answer life's big questions.
Taken together, while moral psychology has primarily focused on how we evaluate specific
actions in specific contexts, our research suggests that people also make judgments about lives
and ways of living as a whole, pointing to the need for a broader framework of human evaluation
concerned with how people evaluate the nature, significance, and direction of life. Understanding
these evaluations not only broadens the scope of moral psychology, but may also help cultivate
better answers to the big questions people face in life: what to value, what to strive for, and
ultimately, how to live.
Reference:
Chen, X., Harris, P. L., & Yang, F. (2023). Beyond enjoyment: Young children consider the
normative goodness of activity engagement when attributing happiness. Journal of
Experimental Child Psychology, 228, 105608. https://doi.org/10.1016/j.jecp.2022.105608
Ge, B. H., & Yang, F. (2023). Transcending the self to transcend suffering. Frontiers in
Psychology, 14. https://doi.org/10.3389/fpsyg.2023.1113965
Huang, M., & Yang, F. (2023). Self-transcendence or self-enhancement: People’s perceptions of
meaning and happiness in relation to the self. Journal of Experimental Psychology:
General, 152(2), 590–610. https://doi.org/10.1037/xge0001297
Kim, Y., & Yang, F. (2026). People value the pursuit of potential-realization more than
belonging, happiness, and status. Self and Identity, 25(1), 93–110.
https://doi.org/10.1080/15298868.2025.2580496
Phillips, J., Misenheimer, L., & Knobe, J. (2011). The Ordinary Concept of Happiness (and
Others Like It). Emotion Review, 3(3), 320–322.
https://doi.org/10.1177/1754073911402385
Phillips, J., De Freitas, J., Mott, C., Gruber, J., & Knobe, J. (2017). True happiness: The role of
morality in the folk concept of happiness. Journal of Experimental Psychology: General,
146(2), 165–181. https://doi.org/10.1037/xge0000252
Soter, L. K., Gelman, S. A., & Yang, F. (2025). What we owe to ourselves: Investigating
people’s sense of obligations to the self. Journal of Personality and Social Psychology.
https://doi.org/10.1037/pspa0000474
Yang, F. (2025). Being good and feeling good: What happiness means to children. Child
Development Perspectives, 19(1), 38–44. https://doi.org/10.1111/cdep.12522
Yang, F., Knobe, J., & Dunham, Y. (2021). Happiness is from the soul: The nature and origins of
our happiness concept. Journal of Experimental Psychology. General, 150(2), 276–288.
https://doi.org/10.1037/xge0000790

Prominent accounts treat moral judgment as contingent on mental-state inference, yet its influence varies across contexts and cultures. We propose a trade-off account: mental-state inference yields benefits (accurate partner assessment, reputational fairness) but incurs costs (effort, delay, uncertainty). Factors modulating these costs and benefits - threat, relationship value, time pressure - should systematically alter reliance on mental states. Four studies (N > 2,000) support this account and suggest a unified explanation of cross-cultural variation in mental-state use.
Full abstract
Prominent accounts of moral judgment treat it as fundamentally contingent on inferring a perpetrator's mental states, their intent, knowledge, and foresight (Cushman, 2008). Yet the influence of mental-state information varies systematically across contexts, individuals, and cultures (Barrett et al., 2016). We propose a trade-off account to explain this variability. Investing in mental-state inference can yield benefits - including more accurate partner selection, the maintenance of valuable relationships, and the preservation of a reputation for fairness - but it also incurs costs: cognitive effort, uncertainty, and delay. Any factor that shifts the relative weight of these costs and benefits should predictably alter how much observers rely on mental-state inference when making moral judgments.
We tested core predictions of this account across four studies using vignettes featuring ambiguous evidence about a wrongdoer's intent, among over 2,000 U.S.-based Prolific participants, with effects evaluated using both frequentist and Bayesian inference. In two studies, participants evaluated the same ambiguous wrongdoing in either a high- or low-threat context, rating (i) how important it was to investigate the actor's mental state prior to punishment, and (ii) how much time it was acceptable to spend doing so. High-threat conditions reduced the perceived importance of mental-state investigation and shortened acceptable deliberation time, consistent with the prediction that swift deterrence takes priority when the costs of delay are elevated. In two further studies, we manipulated the social value of both wrongdoers and victims, measuring how participants inferred intent from identical ambiguous evidence. Acts were perceived as less intentional when the wrongdoer was more socially valued, and more intentional when the victim was more socially valued, consistent with the prediction that relationship-management goals shape the inference process itself.
These findings challenge the view that moral judgment is uniformly or obligatorily mind-sensitive. Rather, mental-state use is facultative: observers invest in mental-state inference when it is instrumentally useful and invest less when its costs are high. This account offers a parsimonious explanation for several well-documented phenomena in moral psychology, including outcome bias (Alicke, 2000), the side-effect effect (Knobe, 2003), moral luck, and the dissociation between mental-state sensitivity in judgments of wrongness, punishment, and character (Cushman, 2008). We further argue that persistent ecological and social factors - including relational mobility and chronic threat - can account for documented cross-cultural variation in mental-state reliance (Barrett et al., 2016), and that the same trade-off logic illuminates why some legal and moral norms approximate strict liability while others remain intent-sensitive.