Moral Perception, Salience, and the Psychology of Ethical Attention

Last Updated May 28, 2026

Moral life does not begin only when people reach a judgment. It begins earlier, at the level of attention. Before a person can condemn cruelty, recognize neglect, respond to suffering, identify exploitation, or notice an obligation, something morally significant must first become visible. This is the domain of moral perception and moral salience: the psychological processes through which harms, vulnerabilities, duties, exclusions, degradations, forms of dependence, and possibilities for care become noticeable rather than remaining background noise.

This matters because moral failure often begins as a failure of attention before it becomes a failure of judgment, motivation, or action. People overlook the vulnerable, normalize the degrading, fail to register indirect harm, avert their eyes from dependency, or attend selectively to some forms of suffering while ignoring others entirely. Institutions do the same at scale. They foreground metrics and hide persons, distribute responsibility until no single wrong is vividly perceived, and structure attention so that some harms remain administratively invisible.

This article argues that moral perception is one of the foundations of ethical agency. People do not encounter a neutral world and then simply add moral judgment afterward. They notice some things and miss others. They see some people as fully vulnerable moral subjects and reduce others to categories, cases, costs, risks, outputs, data points, strangers, or problems. Ethical life depends not only on what people think, but on what they are able—or willing—to see.

The problem of moral perception is especially important in complex societies where harm is often slow, distributed, bureaucratic, statistical, technical, or hidden behind routine. A child crying in front of us may become morally vivid almost immediately. But predatory lending, environmental poisoning, algorithmic exclusion, workplace degradation, medical neglect, educational tracking, housing discrimination, or administrative abandonment may remain less perceptible because no single scene captures the whole injury. Ethical attention must therefore be cultivated, not assumed.

Moral psychology becomes more serious when it asks not only why people judge as they do, but why some realities become morally visible while others disappear. What makes a suffering person noticeable? What makes an institution’s procedure appear neutral rather than harmful? What makes a group’s humiliation appear ordinary? What makes one person’s pain urgent and another person’s pain abstract? These are questions about perception, salience, attention, framing, emotion, power, and social formation. They are also questions about justice.

What Moral Perception Is

Moral perception refers to the way morally relevant features of the world are noticed, registered, and interpreted as morally significant. It concerns the perception of harm, need, disrespect, domination, cruelty, exclusion, neglect, dignity, dependency, vulnerability, obligation, or possibility for repair before these are fully theorized in explicit language. The term does not require the claim that morality is perceived in exactly the same way as color, motion, or shape. It is enough to say that human attention is often attuned—or poorly attuned—to morally relevant features and that this attunement shapes what enters consciousness as ethically important.

Moral perception is not identical with moral judgment. Judgment asks whether something is wrong, right, permissible, required, blameworthy, admirable, or unjust. Perception asks whether the morally relevant feature became available to judgment in the first place. The person who does not notice another’s fear, dependence, exhaustion, humiliation, silence, or exclusion may never reach a moral verdict because the case never becomes a case. It remains ordinary scenery.

Moral salience is closely related but slightly broader. It refers to the degree to which a feature of a situation stands out as morally charged, urgent, or significant. Salience is not a property of the event alone. It is relational. A situation becomes salient to someone under conditions shaped by attention, history, identity, emotion, expectation, role, social position, and institutional framing. What is glaringly unjust to one person may be routine or nearly invisible to another.

Moral perception is therefore a psychological, social, and political problem. It is psychological because attention, appraisal, emotion, memory, and interpretation shape what becomes focal. It is social because communities teach people what to notice and what to ignore. It is political because power shapes visibility: some injuries are named immediately, while others are normalized, hidden, individualized, or explained away. A serious moral psychology must therefore ask not only what is judged, but how moral notice is distributed.

Back to top ↑

Why Ethical Attention Matters

Ethical attention matters because perception sets the conditions of later moral life. People cannot respond to what they have not registered, cannot repair what they have not recognized, and cannot judge fairly if morally relevant features never become focal in the first place. Many ethical failures are not failures of principle in the abstract. They are failures of seeing: the failure to notice suffering, the failure to detect one’s own power, the failure to interpret institutional procedure as harmful, the failure to recognize another as fully vulnerable or fully human.

This is why attention is morally consequential even before it becomes overt choice. It helps determine whose pain counts, which harms are treated as exceptional, which burdens are normalized, and which duties become vivid enough to compete with convenience, ideology, role expectation, professional habit, or fear. The psychology of ethical attention therefore links cognition to justice. It asks how moral life is made possible or impossible by the selective organization of awareness.

Ethical attention also matters because moral life is finite. No person can notice everything with equal intensity. Attention is limited, contested, and shaped by fatigue, distraction, media, work demands, social pressures, and institutional cues. This means the moral question is not simply “Are people attentive?” but “What are they trained to attend to, and at whose expense?” A society’s moral life can be understood partly by studying where its attention repeatedly goes and where it repeatedly fails to go.

At the personal level, ethical attention helps people resist indifference. At the relational level, it makes care possible by registering the needs, boundaries, and vulnerabilities of others. At the institutional level, it helps prevent harm from disappearing into process. At the civic level, it determines which suffering becomes public, which injustice becomes politically speakable, and which communities remain unseen until crisis forces recognition.

Back to top ↑

Salience Before Judgment

Moral judgment presupposes moral salience. Before an observer can decide that an action is cruel, negligent, exploitative, generous, cowardly, admirable, or unjust, something about the situation must first stand out as morally relevant. A vulnerable face, an asymmetry of power, a broken promise, a voice being ignored, an expression of distress, an act of care, or the absence of care may all function as attentional anchors. Once these features are noticed, later judgment becomes possible. When they are not noticed, judgment never fully gets off the ground.

This means that moral life begins earlier than verdict. People often imagine themselves as making judgments from a neutral observational position, but perception is already selective, interpretive, and value-laden. The world does not present itself as equally vivid in all respects. Attention highlights some features and backgrounds others. Ethical life is shaped at that stage of selection.

This also means that moral disagreement is sometimes disagreement about salience before it is disagreement about principle. Two people may not disagree because one rejects the value of dignity or fairness. They may disagree because one sees humiliation where the other sees procedure, one sees exploitation where the other sees efficiency, one sees abandonment where the other sees personal responsibility, one sees domination where the other sees tradition, or one sees need where the other sees inconvenience.

Salience can also explain why people revise moral judgments after new forms of visibility. A testimony, photograph, report, statistic, story, audit, investigation, conversation, or direct encounter can alter what the person perceives as morally relevant. The ethical world changes not because a new rule appears, but because something formerly backgrounded becomes focal. Moral growth often begins with a change in what one can no longer fail to notice.

Back to top ↑

Harm, Vulnerability, and Visibility

One major axis of moral salience is harm. People are often especially responsive when injury, pain, fear, domination, humiliation, or direct vulnerability becomes visible. But harm is not always obvious. Some harms are immediate and embodied; others are slow, distributed, statistical, bureaucratic, environmental, algorithmic, financial, institutional, or socially normalized. A child crying in front of us may be morally vivid in a way that predatory lending, environmental poisoning, algorithmic exclusion, wage theft, housing discrimination, medical neglect, or administrative abandonment is not, even when the latter generate wider injury.

This distinction matters because moral attention can be calibrated to the dramatic and immediate while remaining weakly responsive to the structural and cumulative. Ethical attention is therefore not merely about sensitivity. It is about what kinds of suffering, exposure, or domination become perceptible under different conditions. A mature moral psychology must explain why some forms of harm command attention while others require deliberate interpretive work to become visible at all.

Vulnerability is similarly uneven. Some vulnerability is visible: illness, injury, grief, displacement, fear, age, exhaustion, or obvious dependency. Other vulnerability is hidden: debt, precarity, trauma, discrimination, silence under threat, inaccessible systems, undocumented status, caregiver burden, chronic illness, cognitive load, or institutional powerlessness. Moral perception often fails when vulnerability does not look like culturally recognized vulnerability.

Visibility is therefore not the same as reality. A harm can be real while remaining socially unseen. A vulnerability can be severe while appearing ordinary to those protected from it. A person can be exposed to danger while being interpreted as troublesome, inefficient, irrational, irresponsible, difficult, noncompliant, or invisible. Ethical attention must learn to perceive not only visible suffering but also the conditions that make suffering hard to see.

Back to top ↑

Attention and the Structure of Moral Worlds

Human beings inhabit moral worlds structured by habits of attention. Families, professions, classes, political groups, religious traditions, organizations, and media environments train people to notice some things quickly and to ignore others. A clinician may notice pain, risk, or vulnerability that a manager overlooks. A labor organizer may notice exploitation that a market enthusiast reads as efficiency. A teacher may notice a child’s withdrawal as distress where a disciplinary system reads defiance. A privileged actor may fail to perceive humiliation or exclusion because the environment is designed around the actor’s own experience as normal.

This means moral perception is never simply a private faculty floating free of history and social formation. It is shaped by scripts, practices, expectations, categories, and institutions that distribute visibility unequally. Ethical attention is therefore part of the politics of social reality. To ask what becomes morally salient is also to ask whose world sets the terms of notice.

Different moral worlds produce different perceptual habits. A profession organized around risk management may notice liability before suffering. A bureaucracy organized around throughput may notice delays before dignity. A political movement organized around threat may notice enemies before vulnerability. A market setting organized around efficiency may notice productivity before exploitation. A family system organized around control may notice obedience before fear. These habits are not merely cognitive. They shape ethical possibility.

Because attention is socially trained, moral growth often requires crossing perceptual worlds. Literature, testimony, fieldwork, historical study, conversation, religious practice, professional humility, community engagement, and serious encounter with marginalized perspectives can expand what becomes visible. The ethical self is not simply the self with better opinions. It is also the self whose field of attention has been widened and corrected.

Back to top ↑

Emotion and Attentional Guidance

Emotion often guides moral attention. Compassion can make suffering vivid. Anger can sharpen awareness of insult, betrayal, domination, or injustice. Disgust can cause certain violations to loom large. Guilt can direct attention toward one’s own wrongdoing or omission. Shame can intensify self-exposure. Elevation can make moral beauty and excellence salient. Fear can narrow attention to threat while screening out wider obligations. In this sense, moral perception is not cold scanning followed by later feeling. Feeling itself helps organize what is noticed and how strongly it matters.

Yet emotional guidance is not always reliable. Emotions can make some moral features hypervisible while distorting or suppressing others. Anger can magnify transgression while flattening complexity. Disgust can attach moral weight to stigma or bodily aversion where justice does not warrant it. Compassion can be intense for vivid individuals and weak for anonymous populations. Shame can turn attention inward toward exposure rather than outward toward repair. Fear can make self-protection appear more urgent than responsibility.

Emotion therefore has a double role in moral attention. It is indispensable because it gives salience force. A purely detached awareness of harm may not move action. But emotion is also risky because it can miscalibrate attention. The task is not to remove emotion from moral perception. It is to train emotion to remain answerable to evidence, proportion, dignity, and the actual lives affected.

A mature moral psychology must therefore treat emotion as attentional infrastructure. It asks which emotions make harm visible, which emotions make persons disappear, which emotions widen concern, which emotions narrow concern, and how emotional climates are shaped by institutions, media, and power. Ethical attention is not emotionless. It is emotionally disciplined.

Back to top ↑

Mind Perception and Moral Status

Moral attention is closely linked to mind perception: the tendency to recognize others as beings with feelings, intentions, inner lives, agency, vulnerability, and susceptibility to suffering. When persons are perceived as full moral patients, their vulnerability becomes more salient. When they are depersonalized, objectified, abstracted, stereotyped, or reduced to categories, moral attention weakens. This is one reason dehumanization matters so much. It does not only alter judgment after the fact; it reshapes what is visible as morally significant from the beginning.

Attention to mind helps explain why some entities become focal in moral concern while others remain morally backgrounded. Perceived agency and perceived patiency affect whether observers interpret an event as one of wrongdoing, victimization, duty, repair, threat, or punishment. Ethical attention is thus partly about who is seen as someone to whom things can be done and someone whose experience counts.

Mind perception also shapes institutional life. A person seen as a “case,” “claim,” “ticket,” “risk,” “unit,” “bed,” “consumer,” “beneficiary,” “defendant,” “student ID,” “headcount,” or “cost center” can become administratively visible while becoming morally less visible. The institution can see the category but not the person. This is one of the ways modern systems can maintain formal attention while weakening ethical attention.

To perceive another mind is not merely to infer that someone has inner states. It is to recognize that their experience has moral weight. That recognition changes the attentional field. The cry becomes suffering, the silence becomes fear, the delay becomes burden, the exclusion becomes humiliation, the statistic becomes a life. Moral perception often begins when abstraction is interrupted by the recognition of another’s reality.

Back to top ↑

Situations, Scripts, and Selective Seeing

Attention is guided by situation perception. People enter scenes with scripts about what kind of situation this is: a workplace, a classroom, a market exchange, a police stop, a family dispute, a hospital intake, a disciplinary hearing, a performance review, a customer complaint, a security threat, a procurement decision, a data-quality issue, or a public controversy. These scripts influence what features appear relevant and what kinds of moral interpretation seem available. If a situation is framed primarily as administrative, technical, strategic, disciplinary, or transactional, morally significant features may be screened off.

This is one reason ordinary wrongdoing can occur without obvious malice. People may be looking directly at harm while not quite seeing it as harm. They may perceive role expectations, deadlines, procedures, costs, reputational risk, performance targets, or threats more vividly than dependency, dignity, coercion, exclusion, or avoidable suffering. Moral salience is therefore not only about sensory notice. It is about interpretive framing.

Scripts are useful because they simplify complex situations. But they are morally dangerous when they become too rigid. A disciplinary script may hide distress. A market script may hide coercion. A security script may hide racialized suspicion. A productivity script may hide exhaustion. A compliance script may hide injustice. A neutrality script may hide power. A family loyalty script may hide abuse. Once a script controls attention, some moral interpretations become easier and others become nearly unavailable.

Ethical attention requires script interruption. It asks: What kind of situation is this, and what might this frame be hiding? Who is vulnerable here? Who has less voice? What harm could be normalized by the role? What would this look like from the position of the person most affected? What if the category is correct but morally incomplete?

Back to top ↑

Institutions and Organized Inattention

Institutions do not merely host moral attention; they organize it. They decide what counts as reportable, measurable, urgent, exceptional, normal, disposable, or outside scope. Bureaucratic systems often distribute responsibility and partition perception so that no one actor sees the whole harm. Metrics can displace persons. Procedure can conceal injustice by translating injury into categories, compliance documents, workflow states, or throughput indicators. The result is not always overt cruelty. It is often structured inattention.

This matters because many modern moral failures are administrative before they are emotional. People may not be especially vicious, but they are placed inside systems that render some harms difficult to perceive. Ethical attention in such contexts requires more than personal kindness. It requires institutional design that restores visibility to what procedure tends to hide.

Organized inattention occurs when institutions make certain realities hard to see. This can happen through fragmented responsibility, narrow metrics, inaccessible reporting channels, risk-avoidant legal language, siloed departments, abstract categories, outsourced harm, technical dashboards, algorithmic opacity, or cultures that punish those who notice inconvenient truths. The institution may appear rational while becoming morally blind.

Institutional attention must therefore be designed. Good systems make relevant harms visible, protect dissenting notice, and create channels through which overlooked concerns can become legible before they become catastrophe. They ask not only whether processes are efficient, but whether they preserve contact with the lives affected by those processes. Ethical institutions are not simply rule-following institutions. They are institutions that can still see.

Back to top ↑

Digital Media and Attention Economies

Digital environments intensify the politics of salience. Platforms reward novelty, outrage, immediacy, conflict, identity signaling, emotional display, and rapid reaction. As a result, some moral events become intensely visible while others disappear almost entirely. Attention is not just uneven; it is engineered. This shapes ethical perception by amplifying vivid moral spectacle and weakening patient attention to chronic, dispersed, or structurally complex harms.

The problem is not only overload. It is distortion. Digital moral life can train people to notice what is emotionally viral rather than what is socially important. Under such conditions, ethical attention becomes reactive, intermittent, and public-facing. People learn to respond to what is framed as urgent by the feed, not necessarily to what is morally most serious. The deeper task is to recover forms of attention stable enough to perceive the morally significant even when it is not spectacular.

Digital salience also changes the relationship between perception and performance. A person may notice harm partly through the question of how it will appear publicly: what can be shared, condemned, liked, quoted, mocked, or used for identity display. This does not mean digital concern is always insincere. But it does mean that moral perception becomes entangled with audience, visibility, speed, and social reward. Ethical attention can become performative when public reaction replaces sustained responsibility.

At the same time, digital systems can reveal harms that were previously hidden. Testimony, documentation, citizen journalism, community archives, public data, and networked solidarity can make ignored suffering visible. Digital attention is therefore morally ambivalent. It can expose, amplify, distort, trivialize, weaponize, or sustain moral concern. The question is not whether digital attention is good or bad in general, but what forms of visibility it creates and what forms of responsibility it supports or undermines.

Back to top ↑

Power, Privilege, and the Distribution of Notice

Moral perception is shaped by power because power affects what one has to notice. People with security, status, institutional protection, or social dominance can often move through environments without perceiving the burdens others must constantly track. Those exposed to danger, exclusion, surveillance, precarity, discrimination, or humiliation may develop heightened attention to cues that privileged observers overlook. This asymmetry is not merely personal sensitivity. It is a social distribution of perceptual burden.

Privilege often works by making some harms invisible to those who benefit from the arrangement. A person may not notice access barriers because buildings, schedules, language, credentials, procedures, and assumptions fit their own body, class, accent, race, education, citizenship, gender, religion, or family structure. A professional may not notice institutional intimidation because the system treats them as credible by default. A manager may not notice workload degradation because the metric looks good. A citizen may not notice the coercive edge of policy because it rarely touches their life.

Marginalized experience can therefore carry epistemic significance. Those who bear a harm often notice features of a situation that dominant observers miss or dismiss. This does not mean every individual perception is automatically complete or beyond critique. It means ethical attention must take seriously the standpoint of those who encounter the consequences of power from below. A moral world that ignores affected voices trains itself not to see.

Power also controls institutional categories. It determines what is recorded, what counts as evidence, whose testimony is credible, what becomes a complaint, what is named as harm, and what is dismissed as ordinary discomfort. Ethical attention becomes more just when the terms of notice are not controlled solely by those least burdened by the system.

Back to top ↑

The Failures of Ethical Attention

Moral failures of attention take many forms. There is outright neglect, in which another’s need never becomes focal. There is selective empathy, in which only familiar, innocent, attractive, proximate, or in-group suffering feels real. There is procedural narrowing, in which technical goals displace human significance. There is ideological filtering, in which people fail to notice harms that their worldview normalizes. There is also moral fatigue, where repeated exposure dulls sensitivity and the previously intolerable becomes ordinary.

These failures matter because they reveal that ethical attention is fragile. Seeing morally is not guaranteed by intelligence, education, or good intentions. It can be narrowed by power, habit, self-protection, fear, role conformity, group identity, bureaucratic process, media framing, and systems that distribute visibility unevenly. Moral psychology must therefore ask not only how people know the good, but how they remain capable of noticing what goodness and harm require them to see.

Some failures of attention are motivated. People may not want to see because seeing would require action, guilt, conflict, sacrifice, or loss of innocence. A manager may avoid noticing exhaustion because recognition would require changing expectations. A citizen may avoid noticing state violence because recognition would threaten national identity. A family member may avoid noticing abuse because recognition would disrupt loyalty. In these cases, inattention protects the self or group from moral demand.

Other failures are systemic rather than individually chosen. A person may be too overloaded, too narrowly trained, too embedded in routine, or too dependent on institutional categories to perceive what is happening. This does not erase responsibility, but it changes the remedy. The response must include redesigned attention, not only moral exhortation.

Back to top ↑

Cultivating Moral Perception

If ethical attention can fail, it can also be cultivated. Part of moral development involves learning to notice dependency, asymmetry, exclusion, degradation, neglect, care, courage, and quiet forms of need that previously escaped awareness. Literature, historical study, field experience, clinical practice, civic engagement, religious reflection, serious conversation, restorative practice, and sustained contact with marginalized perspectives can all expand moral salience by widening the range of lives and situations one can perceive with seriousness.

Cultivating moral perception does not mean becoming overwhelmed by every possible harm. It means becoming more accurate, more disciplined, and more responsive in what one notices. The goal is not infinite sensitivity without structure. It is better moral calibration: noticing the morally relevant, interpreting it responsibly, and connecting attention to action where action is possible.

Personal cultivation requires practices of slowing down, listening, perspective-taking, and asking what a situation looks like from the standpoint of those most affected. It also requires attention to one’s own habits of avoidance. What do I quickly dismiss? Whose pain do I explain away? What kinds of harm do I treat as normal? Where do I notice offense to myself more quickly than injury to others?

Institutional cultivation matters too. Good systems make relevant harms visible, protect dissenting notice, and create channels through which overlooked concerns can become legible before they become catastrophe. Ethical attention is therefore both a virtue of persons and a property of environments. A society that wants better moral judgment must also build better conditions of seeing.

Back to top ↑

Mathematical Lens: Modeling Moral Salience

Moral salience can be modeled as the probability that a morally relevant feature of a situation becomes focal enough to influence later evaluation and action. Let \(S_i\) represent the moral salience of event \(i\). A minimal formulation is:

where \(H_i\) is perceived harm, \(V_i\) is visible vulnerability, \(E_i\) is emotional activation, \(M_i\) is mind perception, \(C_i\) is contextual framing, and \(A_i\) is attentional competition from other cues, demands, risks, or distractions.

A more explicit latent-score model is:

The probability that a morally relevant feature becomes focal can then be written as:

Downstream judgment can be modeled as conditional on attention:

Institutions can be added to the model by treating organized inattention as a structural penalty:

Back to top ↑

R Workflow: Modeling Moral Salience and Ethical Attention

The following R workflow simulates data on harm visibility, vulnerability, mind perception, contextual framing, emotional activation, organized inattention, and attentional competition. It estimates which factors make morally relevant features focal and how that focality influences later moral judgment. The dataset is synthetic and intended for reproducible article support, not empirical claims about real persons, employees, students, organizations, institutions, communities, cultures, political groups, or moral worth.

# Moral Perception, Salience, and the Psychology of Ethical Attention
# Synthetic R workflow for modeling moral salience and ethical attention.
# Educational and reproducible research scaffold only.

suppressPackageStartupMessages({
  library(tidyverse)
  library(broom)
})

set.seed(42)

# ------------------------------------------------------------
# 1. Set up output folders
# ------------------------------------------------------------

dir.create("outputs", showWarnings = FALSE)
dir.create("outputs/tables", recursive = TRUE, showWarnings = FALSE)
dir.create("outputs/figures", recursive = TRUE, showWarnings = FALSE)

# ------------------------------------------------------------
# 2. Simulate moral salience data
# ------------------------------------------------------------

n <- 2400

df <- tibble(
  case_id = 1:n,
  perceived_harm = rnorm(n, 0, 1),
  visible_vulnerability = rnorm(n, 0, 1),
  emotional_activation = rnorm(n, 0, 1),
  mind_perception = rnorm(n, 0, 1),
  contextual_framing = rnorm(n, 0, 1),
  attentional_competition = rnorm(n, 0, 1),
  organized_inattention = rnorm(n, 0, 1),
  intention_attribution = rnorm(n, 0, 1),
  reflective_interpretation = rnorm(n, 0, 1),
  repair_pathway = rnorm(n, 0, 1)
) %>%
  mutate(
    latent_salience =
      0.60 * perceived_harm +
      0.50 * visible_vulnerability +
      0.35 * emotional_activation +
      0.40 * mind_perception +
      0.25 * contextual_framing -
      0.45 * attentional_competition -
      0.35 * organized_inattention +
      rnorm(n, 0, 0.8),

    focal_attention_probability = plogis(latent_salience),
    focal_attention = if_else(focal_attention_probability >= 0.5, 1, 0),

    moral_judgment =
      0.70 * focal_attention +
      0.35 * intention_attribution +
      0.30 * reflective_interpretation -
      0.15 * organized_inattention +
      rnorm(n, 0, 0.9),

    ethical_response_tendency =
      0.45 * focal_attention +
      0.30 * moral_judgment +
      0.25 * repair_pathway -
      0.20 * attentional_competition -
      0.20 * organized_inattention +
      rnorm(n, 0, 0.9),

    salience_band = case_when(
      latent_salience < -0.75 ~ "Low moral salience",
      latent_salience < 0.25 ~ "Moderate moral salience",
      latent_salience < 1.0 ~ "High moral salience",
      TRUE ~ "Very high moral salience"
    )
  )

# ------------------------------------------------------------
# 3. Estimate salience-to-attention model
# ------------------------------------------------------------

model_attention <- glm(
  focal_attention ~ perceived_harm + visible_vulnerability +
    emotional_activation + mind_perception +
    contextual_framing + attentional_competition +
    organized_inattention,
  data = df,
  family = binomial()
)

attention_results <- tidy(
  model_attention,
  conf.int = TRUE,
  exponentiate = TRUE
)

attention_fit <- glance(model_attention)

print(attention_results)

# ------------------------------------------------------------
# 4. Estimate judgment model conditional on attention
# ------------------------------------------------------------

model_judgment <- lm(
  moral_judgment ~ focal_attention +
    intention_attribution + reflective_interpretation +
    organized_inattention,
  data = df
)

judgment_results <- tidy(model_judgment, conf.int = TRUE)
judgment_fit <- glance(model_judgment)

print(judgment_results)

# ------------------------------------------------------------
# 5. Estimate ethical response model
# ------------------------------------------------------------

model_response <- lm(
  ethical_response_tendency ~ focal_attention +
    moral_judgment + repair_pathway +
    attentional_competition + organized_inattention,
  data = df
)

response_results <- tidy(model_response, conf.int = TRUE)
response_fit <- glance(model_response)

print(response_results)

# ------------------------------------------------------------
# 6. Summarize by moral salience band
# ------------------------------------------------------------

salience_summary <- df %>%
  group_by(salience_band) %>%
  summarize(
    mean_harm = mean(perceived_harm),
    mean_vulnerability = mean(visible_vulnerability),
    mean_mind_perception = mean(mind_perception),
    mean_competition = mean(attentional_competition),
    mean_organized_inattention = mean(organized_inattention),
    focal_attention_rate = mean(focal_attention),
    mean_judgment = mean(moral_judgment),
    mean_response = mean(ethical_response_tendency),
    .groups = "drop"
  )

print(salience_summary)

# ------------------------------------------------------------
# 7. Build interpretation grid
# ------------------------------------------------------------

pred_grid <- expand_grid(
  perceived_harm = seq(-2, 2, length.out = 120),
  visible_vulnerability = c(-1, 0, 1),
  emotional_activation = 0,
  mind_perception = 0,
  contextual_framing = 0,
  attentional_competition = 0,
  organized_inattention = 0
)

pred_grid$predicted_attention <- predict(
  model_attention,
  newdata = pred_grid,
  type = "response"
)

pred_grid <- pred_grid %>%
  mutate(
    vulnerability_label = case_when(
      visible_vulnerability == -1 ~ "Low visible vulnerability",
      visible_vulnerability == 0 ~ "Average visible vulnerability",
      TRUE ~ "High visible vulnerability"
    )
  )

# ------------------------------------------------------------
# 8. Plot moral salience and focal attention
# ------------------------------------------------------------

plot_attention <- ggplot(
  pred_grid,
  aes(x = perceived_harm, y = predicted_attention)
) +
  geom_line(linewidth = 1) +
  facet_wrap(~ vulnerability_label) +
  labs(
    title = "Predicted Focal Moral Attention",
    subtitle = "Harm and visible vulnerability shape whether moral features become salient",
    x = "Perceived harm",
    y = "Probability of focal attention"
  ) +
  theme_minimal(base_size = 12)

print(plot_attention)

# ------------------------------------------------------------
# 9. Export outputs
# ------------------------------------------------------------

write_csv(df, "outputs/tables/moral_salience_simulated_data.csv")
write_csv(attention_results, "outputs/tables/moral_salience_attention_model.csv")
write_csv(attention_fit, "outputs/tables/moral_salience_attention_model_fit.csv")
write_csv(judgment_results, "outputs/tables/moral_salience_judgment_model.csv")
write_csv(judgment_fit, "outputs/tables/moral_salience_judgment_model_fit.csv")
write_csv(response_results, "outputs/tables/moral_salience_response_model.csv")
write_csv(response_fit, "outputs/tables/moral_salience_response_model_fit.csv")
write_csv(salience_summary, "outputs/tables/moral_salience_summary.csv")
write_csv(pred_grid, "outputs/tables/moral_salience_prediction_grid.csv")

ggsave(
  filename = "outputs/figures/predicted_focal_moral_attention.png",
  plot = plot_attention,
  width = 10,
  height = 6,
  dpi = 300
)

This workflow is useful because it formalizes the article’s main claim: moral judgment is partly downstream of whether a morally relevant feature became focal in the first place. It also adds organized inattention as an institutional factor, making visible how systems can reduce moral salience even when harm and vulnerability are present.

Back to top ↑

Python Workflow: Simulating Ethical Attention Under Competing Salience Cues

The Python workflow below simulates moral salience under conditions of competing attentional demands. It models how harm, vulnerability, mind perception, and contextual framing increase ethical notice, while distraction, organized inattention, and competing cues reduce it. The example uses synthetic data for reproducible demonstration and should not be interpreted as an assessment of real people, institutions, organizations, communities, cultures, political groups, or moral worth.

# Moral Perception, Salience, and the Psychology of Ethical Attention
# Python workflow for synthetic ethical-attention modeling.
# Educational and reproducible research scaffold only.

from pathlib import Path

import numpy as np
import pandas as pd

np.random.seed(42)

# ------------------------------------------------------------
# 1. Set up output folders
# ------------------------------------------------------------

output_tables = Path("outputs/tables")
output_tables.mkdir(parents=True, exist_ok=True)

# ------------------------------------------------------------
# 2. Simulate ethical attention data
# ------------------------------------------------------------

n = 2600

df = pd.DataFrame({
    "case_id": np.arange(1, n + 1),
    "perceived_harm": np.random.normal(0, 1, n),
    "visible_vulnerability": np.random.normal(0, 1, n),
    "emotional_activation": np.random.normal(0, 1, n),
    "mind_perception": np.random.normal(0, 1, n),
    "contextual_framing": np.random.normal(0, 1, n),
    "attentional_competition": np.random.normal(0, 1, n),
    "organized_inattention": np.random.normal(0, 1, n),
    "intention_attribution": np.random.normal(0, 1, n),
    "reflective_interpretation": np.random.normal(0, 1, n),
    "repair_pathway": np.random.normal(0, 1, n)
})

# ------------------------------------------------------------
# 3. Compute latent salience and focal attention
# ------------------------------------------------------------

latent_salience = (
    0.60 * df["perceived_harm"] +
    0.50 * df["visible_vulnerability"] +
    0.35 * df["emotional_activation"] +
    0.40 * df["mind_perception"] +
    0.25 * df["contextual_framing"] -
    0.45 * df["attentional_competition"] -
    0.35 * df["organized_inattention"] +
    np.random.normal(0, 0.8, n)
)

df["latent_salience"] = latent_salience
df["focal_attention_probability"] = 1 / (1 + np.exp(-latent_salience))
df["focal_attention"] = (df["focal_attention_probability"] >= 0.5).astype(int)

# ------------------------------------------------------------
# 4. Generate downstream moral judgment and response
# ------------------------------------------------------------

df["moral_judgment"] = (
    0.70 * df["focal_attention"] +
    0.35 * df["intention_attribution"] +
    0.30 * df["reflective_interpretation"] -
    0.15 * df["organized_inattention"] +
    np.random.normal(0, 0.9, n)
)

df["ethical_response_tendency"] = (
    0.45 * df["focal_attention"] +
    0.30 * df["moral_judgment"] +
    0.25 * df["repair_pathway"] -
    0.20 * df["attentional_competition"] -
    0.20 * df["organized_inattention"] +
    np.random.normal(0, 0.9, n)
)

# ------------------------------------------------------------
# 5. Summarize by attention level
# ------------------------------------------------------------

summary = (
    df.groupby("focal_attention")
      .agg(
          mean_harm=("perceived_harm", "mean"),
          mean_vulnerability=("visible_vulnerability", "mean"),
          mean_emotion=("emotional_activation", "mean"),
          mean_mind_perception=("mind_perception", "mean"),
          mean_competition=("attentional_competition", "mean"),
          mean_organized_inattention=("organized_inattention", "mean"),
          mean_judgment=("moral_judgment", "mean"),
          mean_response=("ethical_response_tendency", "mean")
      )
      .reset_index()
)

print(summary)

# ------------------------------------------------------------
# 6. Create scenario grid for interpretation
# ------------------------------------------------------------

scenario_rows = []

for harm in np.linspace(-2, 2, 41):
    for vulnerability in [-1, 0, 1]:
        for organized_inattention in [-1, 0, 1]:
            latent = (
                0.60 * harm +
                0.50 * vulnerability +
                0.35 * 0 +
                0.40 * 0 +
                0.25 * 0 -
                0.45 * 0 -
                0.35 * organized_inattention
            )
            probability = 1 / (1 + np.exp(-latent))

            scenario_rows.append({
                "perceived_harm": harm,
                "visible_vulnerability": vulnerability,
                "organized_inattention": organized_inattention,
                "predicted_focal_attention_probability": probability
            })

scenario_df = pd.DataFrame(scenario_rows)

print(scenario_df.head(12))

# ------------------------------------------------------------
# 7. Identify low-attention high-harm synthetic cases
# ------------------------------------------------------------

low_attention_high_harm = (
    df[
        (df["perceived_harm"] > df["perceived_harm"].quantile(0.75)) &
        (df["focal_attention"] == 0)
    ]
    .sort_values(
        ["organized_inattention", "attentional_competition"],
        ascending=False
    )
    .head(25)
    .reset_index(drop=True)
)

# ------------------------------------------------------------
# 8. Identify high-salience high-response synthetic cases
# ------------------------------------------------------------

high_salience_high_response = (
    df[
        (df["latent_salience"] > df["latent_salience"].quantile(0.75)) &
        (df["ethical_response_tendency"] > df["ethical_response_tendency"].quantile(0.75))
    ]
    .sort_values(
        ["repair_pathway", "mind_perception"],
        ascending=False
    )
    .head(25)
    .reset_index(drop=True)
)

# ------------------------------------------------------------
# 9. Export files
# ------------------------------------------------------------

df.to_csv(output_tables / "ethical_attention_python_simulation.csv", index=False)
summary.to_csv(output_tables / "ethical_attention_summary.csv", index=False)
scenario_df.to_csv(output_tables / "ethical_attention_scenarios.csv", index=False)
low_attention_high_harm.to_csv(
    output_tables / "ethical_attention_low_attention_high_harm_cases.csv",
    index=False
)
high_salience_high_response.to_csv(
    output_tables / "ethical_attention_high_salience_high_response_cases.csv",
    index=False
)

print("Synthetic ethical-attention outputs written to:", output_tables)

This workflow is useful because it makes ethical attention analyzable as a structured process rather than treating moral notice as mysterious or automatic. It also allows readers to inspect the kinds of cases most central to the article’s argument: high-harm situations that fail to become focal under organized inattention, and high-salience situations that become more likely to produce judgment and response when repair pathways are available.

In a full article repository, this Python workflow can be extended into notebooks, SQL schema, synthetic datasets, validation notes, low-attention high-harm case analysis, organized-inattention simulations, digital-salience scenarios, institutional ethical-attention audits, and additional language examples. R can support statistical modeling and visualization; Python can support simulation and data pipelines; SQL can preserve structured scenario metadata; Julia can support dynamic salience models; and C, C++, Fortran, Go, and Rust can support reproducible command-line tools, validation utilities, and computational demonstrations.

Back to top ↑

GitHub Repository

The companion repository for this article provides a reproducible code scaffold for modeling moral perception, moral salience, ethical attention, perceived harm, visible vulnerability, emotional activation, mind perception, contextual framing, attentional competition, organized inattention, focal attention, moral judgment, repair pathways, ethical response tendency, low-attention high-harm cases, and high-salience high-response cases.

The repository structure should support a full research workflow rather than a single script. The article folder can include language-specific examples in python, r, julia, sql, c, cpp, fortran, go, and rust, along with data, docs, notebooks, and outputs. This structure makes the article reproducible, inspectable, and extensible for readers who want to move from conceptual argument to analytical demonstration.

Back to top ↑

Conclusion

Moral perception and moral salience show that ethical life begins before explicit verdict. Human beings do not first inhabit a neutral visual field and then add morality afterward. They notice some harms and miss others, recognize some vulnerabilities and normalize others, perceive some persons as full moral subjects and background others altogether. The psychology of ethical attention is therefore not peripheral to morality. It is one of its foundations.

To understand moral agency well is to understand how attention is organized: by emotion, by scripts, by institutions, by power, by media, by training, by fatigue, and by cultivated forms of notice. Moral judgment matters, but it is always partly downstream of what became visible enough to matter. A serious ethics of human life must therefore ask not only what people believe or decide, but what their worlds have trained them to see.

This is also why moral perception is a justice issue. Social life does not distribute visibility evenly. Some people’s suffering becomes urgent quickly; others must struggle to be believed. Some injuries are legible to institutions; others disappear into paperwork, delay, metrics, or silence. Some lives are perceived as fully minded; others are reduced to categories, threats, burdens, or abstractions. Ethical attention must therefore be trained against the defaults of power.

The work of moral perception is not simply to see more, but to see more truthfully. It is to notice harm without turning it into spectacle, to recognize vulnerability without reducing persons to need, to see structural injury without losing sight of individual lives, and to build institutions capable of noticing what their own procedures tend to hide. Moral judgment begins when something comes into view. Moral responsibility begins when we ask why so much remains unseen.

Back to top ↑

Related articles

• What Is Moral Psychology?
• Moral Judgment and the Psychology of Right and Wrong
• Intuition, Reflection, and the Structure of Moral Evaluation
• Moral Emotion: Guilt, Shame, Disgust, Compassion, and Elevation
• Conscience, Guilt, Shame, and Moral Self-Evaluation
• Care, Empathy, and Relational Moral Life
• Moral Disengagement and the Psychology of Ethical Failure
• Moral Psychology in Organizations and Institutions

Further reading

• Doris, J. (2006, rev. 2022) ‘Moral Psychology: Empirical Approaches’, in Zalta, E.N. and Nodelman, U. (eds.) The Stanford Encyclopedia of Philosophy. Available at: https://plato.stanford.edu/entries/moral-psych-emp/.
• Gantman, A.P. and Van Bavel, J.J. (2015) ‘Moral Perception’, Trends in Cognitive Sciences, 19(11), pp. 631–633. Available at: https://www.sciencedirect.com/science/article/abs/pii/S1364661315001813.
• Vance, J. and Werner, P. (2022) ‘Attentional Moral Perception’, Journal of Moral Philosophy. Available at: https://philpapers.org/archive/VANAMP-8.pdf.
• Werner, P. (2020) ‘Moral Perception’, Philosophy Compass, 15(2). Available at: https://compass.onlinelibrary.wiley.com/doi/10.1111/phc3.12640.
• Gray, K. and Wegner, D.M. (2012) ‘Mind Perception Is the Essence of Morality’, Psychological Inquiry, 23(2), pp. 101–124. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC3379786/.
• Brezina, T., Agnew, R., Cullen, F.T. and Wright, J.P. (2024) ‘Moral Salience, Situational Moral Evaluations, and Criminal Decision-Making’, Crime & Delinquency. Available at: https://journals.sagepub.com/doi/10.1177/00111287241259474.
• Hutton, J. (2023) ‘What Attentional Moral Perception Cannot Do but Emotions Can’, Philosophies, 8(6), 106. Available at: https://www.mdpi.com/2409-9287/8/6/106.
• Murdoch, I. (1970) The Sovereignty of Good. London: Routledge.
• Diamond, C. (1991) The Realistic Spirit: Wittgenstein, Philosophy, and the Mind. Cambridge, MA: MIT Press.
• Nussbaum, M.C. (1990) Love’s Knowledge: Essays on Philosophy and Literature. New York: Oxford University Press.

Back to top ↑

References

• Brezina, T., Agnew, R., Cullen, F.T. and Wright, J.P. (2024) ‘Moral Salience, Situational Moral Evaluations, and Criminal Decision-Making’, Crime & Delinquency. Available at: https://journals.sagepub.com/doi/10.1177/00111287241259474.
• Diamond, C. (1991) The Realistic Spirit: Wittgenstein, Philosophy, and the Mind. Cambridge, MA: MIT Press.
• Doris, J. (2006, rev. 2022) ‘Moral Psychology: Empirical Approaches’, in Zalta, E.N. and Nodelman, U. (eds.) The Stanford Encyclopedia of Philosophy. Available at: https://plato.stanford.edu/entries/moral-psych-emp/.
• Gantman, A.P. and Van Bavel, J.J. (2015) ‘Moral Perception’, Trends in Cognitive Sciences, 19(11), pp. 631–633. Available at: https://www.sciencedirect.com/science/article/abs/pii/S1364661315001813.
• Gray, K. and Wegner, D.M. (2012) ‘Mind Perception Is the Essence of Morality’, Psychological Inquiry, 23(2), pp. 101–124. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC3379786/.
• Hutton, J. (2023) ‘What Attentional Moral Perception Cannot Do but Emotions Can’, Philosophies, 8(6), 106. Available at: https://www.mdpi.com/2409-9287/8/6/106.
• Murdoch, I. (1970) The Sovereignty of Good. London: Routledge.
• Nussbaum, M.C. (1990) Love’s Knowledge: Essays on Philosophy and Literature. New York: Oxford University Press.
• Vance, J. and Werner, P. (2022) ‘Attentional Moral Perception’, Journal of Moral Philosophy. Available at: https://philpapers.org/archive/VANAMP-8.pdf.
• Werner, P. (2020) ‘Moral Perception’, Philosophy Compass, 15(2). Available at: https://compass.onlinelibrary.wiley.com/doi/10.1111/phc3.12640.

Back to top ↑