The Illusion of Moral Decline

At a Glance

1. Quick Summary

Ask almost anyone, in almost any country, whether people are as honest and considerate as they were twenty years ago, and the answer is no. Adam Mastroianni and Daniel Gilbert tested that answer in a 2023 paper in Nature, drawing on survey responses from roughly 12.5 million people across 60 nations and 70 years. On 84 to 86 percent of the items they examined, a majority said morality was getting worse. Yet surveys asking people how others were treating them right now stayed flat across the same decades. People also date the start of the decline to roughly the year they were born, and report no decline among the people they know. The likeliest explanation is a pair of ordinary habits: we see far more bad news about strangers than good, and our bad memories fade faster than our good ones.

Simply About the Bias

The illusion of moral decline is the feeling that people used to be nicer, more honest and more fair than they are now. Almost everyone feels it. The strange part is that when researchers check, people today say strangers treat them about as well as people said fifty years ago. Nothing got worse. It just feels that way.

Think about your school. An older cousin tells you kids were way kinder when she was there and nobody was mean online. But she never saw the mean notes passed around back then, and the ones she did see, she has mostly forgotten. Meanwhile you see every nasty comment on your phone the second it appears. The bad stuff from her time faded. The bad stuff from your time is right in front of you. So her school looks friendlier, even if it wasn't.

Here is a trick for catching it. When you think "people used to be better," ask one question: are the people I actually know getting worse? Your friends, your family, your teachers? Usually not. If the people you can see are fine and only the strangers on the news seem terrible, you are looking at the illusion.

2. The Science Behind It

2.1. Discovery and History

The complaint is ancient. Around 20 BC the Roman poet Horace wrote that his parents' generation was worse than his grandparents', and that his own would raise children more vicious yet. A few years later Livy opened his history of Rome by describing how morals "first subsided, as it were, then sank lower and lower." Each generation since has located the golden age about one lifetime back, where its own childhood sits.

Psychologists long treated this as nostalgia or as a feature of conservative politics. Richard Eibach, Lisa Libby and Thomas Gilovich moved it toward the lab in 2003 with a paper arguing that people mistake change in themselves for change in the world: becoming a parent makes a person more alert to threats, and the new alertness gets read as a rise in actual danger. John Protzko and Jonathan Schooler added "Kids these days" in 2019: adults who excel at a trait are the ones most convinced that today's young people lack it.

The decisive step came in 2023. Adam Mastroianni, then a postdoctoral researcher at Columbia, and Daniel Gilbert of Harvard assembled every survey item they could find, from 1949 onward, that asked whether morality was getting better or worse. Then they set those beliefs beside a different archive of surveys that asked about present-day moral behaviour, and checked whether the second series moved. It did not. "The illusion of moral decline" appeared in Nature in June 2023, and later that year Transparent Replications, a project of Clearer Thinking, reran Study 5b with a larger sample and found all of its primary results again.

The caveats: the experiments used American participants, and the mechanism the authors propose, biased exposure plus biased memory, is a well-supported candidate rather than a proven cause, as the paper itself says. Neither touches the central finding. The belief is near universal, and the behavioural measures it rests on have not moved.

Key milestones:

• c. 27 BC: Livy's preface describes Roman morals sinking "lower and lower"
• 2003: Eibach, Libby and Gilovich show that change in the self is mistaken for change in the world
• 2019: Protzko and Schooler publish "Kids these days" in Science Advances
• 2023: Mastroianni and Gilbert publish "The illusion of moral decline" in Nature
• 2023: Transparent Replications reproduces all primary findings of Study 5b

2.2. Key Researchers

2.3. Landmark Studies

"The illusion of moral decline" (Mastroianni & Gilbert, 2023)

The archival section gathered 177 survey items put to 220,772 Americans between 1949 and 2019, each asking whether people were becoming less honest, kind or moral. A majority reported decline on 84 percent of them. A second archive covered 58 items answered by 354,120 people in 59 other nations between 1996 and 2007; a majority reported decline on 86 percent. In Pew surveys of 40 countries in 2002 and 2006, a majority in every country called moral decline at least a moderately big problem.

The authors then asked whether moral behaviour, as people reported experiencing it, had changed. They gathered 107 items answered by roughly 4.5 million people between 1965 and 2020: whether respondents had been treated with respect that day, whether most people could be trusted, whether a stranger had been kind to them. Survey year explained less than one percent of the variation in almost every case.

The experiments sharpened the picture. Participants rated people as less kind, honest, nice and good in 2020 than in 2005. Asked to rate morality in a series of years from before their birth to the present, they reported no decline before they were born, then steady decline from around the time they arrived, at the same rate per year whatever their age. Asked about the people in their own lives rather than people in general, they reported slight improvement.

The Transparent Replications test (2023)

Transparent Replications, which independently reruns studies from top journals, replicated Study 5b with 533 participants, against 387 in the original. Every primary finding reproduced: participants again perceived decline over their own lifetimes and placed its start near their birth. The project gave the study its top rating for replicability, with a lower score on transparency because the original was not preregistered.

"Kids these days" (Protzko & Schooler, 2019)

Across 3,458 American adults, the more authoritarian a respondent, the more they believed children today disrespect their elders; the more well-read, the more they believed children no longer enjoy reading. When participants were led to believe they were less well-read than they thought, the complaint softened.

2.4. Neurological Basis

Nobody has scanned the brain of a person in the act of believing that morality is declining; the topic is too new for a dedicated neural literature. What can be said rests on the two biases that feed it.

The first is the negativity bias. Event-related potential work from John Cacioppo's laboratory in the late 1990s showed that negative stimuli produce larger evaluative brain responses than equally intense positive ones, so bad information is weighted more heavily at encoding. The second is the fading affect bias, reviewed by Walker, Skowronski and Thompson in 2003: the emotional charge of unpleasant memories fades faster than that of pleasant ones. Its neural mechanism is debated, but the pattern holds across cultures and ages and is weaker in people with depression, which suggests an active regulatory process rather than simple decay.

Put the two together and you have what Mastroianni and Gilbert call BEAM, for biased exposure and memory. There is no moral decline circuit, only an ordinary intake bias meeting an ordinary storage bias, both pointing the same way.

3. Evolutionary Origins

Each component of the illusion was useful on its own; the illusion is a side effect of their combination.

Weighting bad news heavily makes sense when the cost of missing a threat is death and the cost of overreacting is a wasted afternoon. Attending to moral violations paid off in small groups where survival depended on cooperation: knowing who cheated was core social intelligence. Selection had no reason to build a mind that balanced each story of betrayal with a story of ordinary decency, because ordinary decency required no action.

Letting bad feelings fade is useful for a different reason. An animal as distressed by last year's injuries as by today's would have little capacity left for the present. The fading affect bias keeps the emotional ledger from being swamped, and there is evidence it supports mental health. The price is a past remembered as kinder than it was.

A third ingredient may be the usefulness of the complaint itself. Declaring that standards are slipping is a way of asserting standards, and it pressures younger members of a group to conform. A belief that flatters the believer, enforces norms and costs nothing to hold will persist regardless of accuracy.

4. How This Bias Manifests

4.1. In Everyday Life

• Telling a younger relative that "people used to look out for each other," while being unable to name a way one's own neighbours have become worse
• Reading one viral video of a rude customer as proof that courtesy is dying, while hundreds of uneventful transactions leave no trace
• Remembering a childhood town as a place where doors were left unlocked, and forgetting the burglaries, bullying and feuds that also happened there

4.2. In the Workplace

• Senior staff insisting that new hires lack the work ethic or manners of their own cohort, a complaint recorded about every generation since offices existed
• Surveillance software introduced to counter a perceived decline in honesty, corroding the trust it is meant to protect
• Interviewers discounting younger candidates on the assumption that integrity is a vanishing trait

4.3. In Business and Marketing

• Advertising that sells products as a return to a more honest era: "made the way things used to be made," "old-fashioned service"
• Security products marketed on the premise that strangers have become more dangerous
• Media businesses serving a steady diet of moral outrage because outrage is what audiences click on

4.4. In Politics and Media

The illusion is a political resource. Mastroianni and Gilbert found that conservative respondents perceived somewhat more decline than liberals, but liberals perceived decline too; the difference was one of degree. A campaign that promises to restore lost decency speaks to a feeling voters already hold. News organisations are the main engine of the exposure bias: a stranger's crime in another city is now as accessible as a neighbour's. The volume of moral failure a person sees in a day has risen enormously; the rate at which strangers misbehave has not.

Specific manifestations include:

• Speeches that treat the perceived decline as a settled fact requiring policy rather than a perception requiring examination
• Moral panics about youth behaviour that recur each decade with a new object (comics, television, video games, phones)

4.5. In Healthcare

• Clinicians who believe patients have become more demanding, when complaint rates in their own practice have not shifted
• Older practitioners discounting trainees because "doctors used to take the vocation seriously"
• Harm from the belief itself: chronic conviction that the social world is deteriorating goes with lower trust and greater anxiety

4.6. In Finance and Investing

• The belief that markets were once run by honourable people and are now full of cheats, despite a documented history of fraud in every century of finance
• Investors holding cash or gold because "nobody can be trusted anymore," paying real returns for a perceived risk that has not changed
• Boards tightening governance in response to a feeling that ethics are slipping, a feeling that existed before every past scandal

5. Real-World Case Studies

Case Study 1: Seventy Years of Polling and the Flat Line Beneath It

• Context: Gallup and other pollsters have asked Americans since the late 1940s whether honesty and morality in the country are getting better or worse.
• What happened: When Mastroianni and Gilbert compiled these items, 177 in all, a majority answered "worse" on 84 percent of them. In 1949, worse; in 1979, worse; in 2019, worse. Meanwhile the General Social Survey had been asking "Can most people be trusted?" across the same decades. Those answers barely moved.
• The bias at work: Each cohort compared the present it could see, saturated with reported wrongdoing, against a past it could only remember, from which the worst had faded.
• Consequences: Seventy years of national conversation, legislation and campaigning premised on a decline the behavioural measures do not show.
• Lessons learned: A belief can be stable, popular and sincere across decades while being wrong. The only test is to put it next to a measure of the thing it claims to describe.

Case Study 2: The Decline That Began the Year You Were Born

• Context: In Study 5b of the Nature paper, later replicated by Transparent Replications, participants rated how kind, honest, nice and good people were in a series of years from before their birth to the present.
• What happened: Participants reported no change in the years before they were born. Decline began around their birth year and continued to the present. A 25-year-old saw the slide start in the late 1990s; a 65-year-old saw it start in the late 1950s. The rate per year was the same for both.
• The bias at work: The years before birth are known only secondhand. The years since birth are remembered, and those memories have faded in the lopsided way the fading affect bias predicts. The result is a decline that tracks the participant's own lifespan rather than any event in the world.
• Consequences: Every person carries a private golden age that coincides with early childhood and treats it as historical fact.
• Lessons learned: When the start of a trend lines up with the observer's own arrival, the trend is probably in the observer.

Historical Example: Horace, Livy and the Rot That Never Finished

Two of Rome's most read authors, writing within a few years of each other in the first century BC, each diagnosed a moral collapse already far advanced. Horace, in his sixth Roman Ode, traced the decline across four generations and predicted it would continue. Livy, in the preface to his history, described morals sinking until Rome could endure "neither our vices nor their cure."

If both were right, Rome was past saving by 20 BC. Yet the Roman state continued for five more centuries in the West and fifteen in the East, and later writers diagnosed fresh declines, each pointing back toward a better age that overlapped with Livy's supposedly rotten one. No real trend could run downward for two thousand years without reaching bottom. The only constant is the complaint.

6. The Cost of This Bias

6.1. Personal Costs

• Lower trust in strangers, which means fewer helpful interactions and a lonelier daily life
• Chronic low-grade anxiety about a social world that feels more hostile than it is
• Jobs, friendships and neighbourhood ties forgone because "people aren't like they used to be"

6.2. Professional Costs

• Generational stereotyping in hiring, mentoring and promotion that discards capable people on the basis of an imagined trend
• Control policies introduced to counter a perceived collapse in honesty that drive away the trustworthy employees they inconvenience most
• Poor forecasting in any field that depends on predicting behaviour, since the forecaster assumes a downward slope the data do not contain

6.3. Societal Costs

• Support for punitive policies justified by a decline that is not occurring, from harsher sentencing to restrictions on young people
• A media economy that profits from outrage, with the illusion supplying the audience and the audience sustaining the illusion
• Difficulty mobilising collective action, because people who believe their neighbours have become selfish do not expect cooperation and so do not offer it

6.4. Statistical Impact

• In the US archive, a majority reported moral decline on 84 percent of 177 survey items covering 1949 to 2019
• In the international archive, a majority reported decline on 86 percent of 58 items across 59 nations, and majorities in all 40 nations in the 2002 and 2006 Pew surveys called it at least a moderately big problem
• In 107 items on contemporaneous behaviour, answered by about 4.5 million people between 1965 and 2020, survey year explained less than one percent of the variation in nearly every case
• The Transparent Replications test of Study 5b (533 participants versus 387) reproduced every primary finding

7. The Hidden Benefits

A belief this widespread and durable is probably doing something for the people who hold it.

• It keeps standards visible: Complaining that honesty is slipping restates that honesty matters. A community in which nobody worried about declining decency might be one in which decency had stopped being discussed.

• It protects well-being: One half of the mechanism, the fading of negative memories, is a feature of healthy emotional regulation. Nobody should want to switch it off in order to see the past more accurately.

• It occasionally flags real problems: Moral behaviour can decline in specific places and periods. The illusion is a bias toward seeing decline, not a guarantee that decline never happens.

These benefits are modest and do not offset the cost of mistrusting the people around you on the basis of a trend that is not there. The useful move is to keep the standards and lose the false history.

8. Self-Assessment: Do You Have This Bias?

8.1. Warning Signs Checklist

• You are confident that people were kinder and more honest when you were young, but cannot point to a measure that shows it
• Most of your evidence for moral decline comes from the news or social media rather than from people you have dealt with directly
• The people you actually know seem as decent as ever, yet you believe "people in general" have gone downhill
• Unpleasant events from your past (bullying, cheating, cruelty you witnessed) come to mind only with effort
• You have said some version of "kids these days" in the last month
• A single viral story of someone behaving badly feels like proof of a trend
• You date the start of the decline to roughly your own childhood
• You assume strangers today would be less likely to return a lost item than strangers in the past
• Stories of ordinary kindness surprise you, as though they were exceptions
• You feel a background sense that the social world is getting worse, without being able to say how

Scoring:

• 0-2 checked: Low susceptibility
• 3-5 checked: Moderate susceptibility
• 6-8 checked: High susceptibility
• 9-10 checked: Very high susceptibility

8.2. Self-Reflection Questions

1. When you picture the "better" past, what year is it, and how old were you? What could you notice about how adults treated each other at that age?
2. In the last week, how many strangers did you see behaving badly in person, as opposed to on a screen?
3. Think of three people you have known for twenty years or more. Have they become less honest or less kind? If not, why would strangers have?
4. What unpleasant things happened in your community during your childhood that you rarely think about now?
5. If someone showed you data that people report being treated as well today as in 1970, what would it take to believe it?

8.3. Quick Diagnostic Scenario

Scenario: At a family gathering, your uncle says people have no decency anymore and cites a news story about a man who filmed himself mocking a homeless person. "That would never have happened when I was young."

How would you respond?

• A) Agree, and add a story of your own about rudeness you saw recently → High susceptibility
• B) Feel he is probably right about the trend but think the example is extreme → Moderate susceptibility
• C) Wonder how many people mocked homeless people in his youth without a camera present, and whether the people he actually knows have changed → Low susceptibility

9. Identifying This Bias in Others

9.1. Behavioral Indicators

• Treats any news story about bad behaviour as evidence of a trend rather than an event
• Describes a past era in uniformly warm terms, with no concrete examples of its cruelties
• Rates the people in their own life as fine while rating "people in general" as deteriorating
• Grows more certain of decline with age, and dates its start to their own early years
• Dismisses survey data on trust or helping as naive or manipulated

9.2. Conversational Red Flags

Phrases people say when under this bias:

• "Nobody has any respect anymore."
• "When I was growing up, you could leave your door unlocked."
• "Kids these days..."
• "That would never have happened back then."

Types of arguments they make:

• A vivid anecdote, usually from the news, standing in for a trend
• A personal memory of a pleasant past offered as historical record

Questions they avoid asking:

• "How would I know what strangers were like in 1975?"
• "Have the people I know actually changed?"

9.3. Situational Triggers

• Exposure to a viral story of cruelty or dishonesty
• Milestones that change one's vantage point: becoming a parent, moving, retiring
• Conversations among people of the same generation, where the decline story is shared and reinforced
• Nostalgic media, from period dramas to "remember when" content, that depicts the past with its unpleasantness removed

10. Cognitive Debiasing Strategies

10.1. Immediate Techniques

• The Personal-World Check: The moment you think "people have gotten worse," name five people you have known for fifteen years and ask whether they have. If your own world is fine, the decline lives in your news feed.
• The Birth-Year Test: Ask when the decline started. If the answer is close to your own childhood, you are measuring your memory, not the world.
• The Camera Question: For any viral example of bad behaviour, ask how often the same thing happened before everyone carried a camera. The behaviour is not new; the footage is.

10.2. Long-Term Strategies

• Keep a Kindness Ledger: For a month, write down every time a stranger does something decent for you or in front of you. The exposure bias is a counting error; the ledger corrects the count.
• Change the Information Diet: Reduce outrage-driven news and social media, to restore a sample of strangers' behaviour that resembles reality.
• Separate Standards from History: Practise saying "I want people to be more honest" instead of "people used to be more honest."

10.3. Environmental Design

• Dashboards over anecdotes: Keep actual measures of conduct (complaint rates, audit findings, trust surveys) visible, so an alarming story can be checked against a trend line within minutes.
• Mixed-age teams: Decline stories thrive in single-cohort groups. Working closely with people of different ages makes it harder to believe another cohort is uniformly worse.
• A rule for viral stories: Agree, at home or at work, that no policy changes in response to a single story until someone has looked at the numbers.

10.4. When to Seek External Input

• When you are about to make a decision (hiring, policy, whom to trust) on the premise that people have become less honest
• When you are certain of a trend but cannot cite anything beyond news stories and personal memory
• When the perceived decline coincides with a change in your own life, such as a new child or a new job

Whom to ask: someone a generation younger or older than you, who has a different golden age and will not share your baseline. Ask "What do the numbers say?" rather than "Don't you think things have gotten worse?"

11. Practical Exercises

Exercise 1: The Two-Column Week

• Objective: Make the exposure bias visible by counting what you actually see
• Time required: 2 minutes, several times a day, for one week
• Materials needed: A notebook or phone note with two columns
• Difficulty level: Beginner
• Instructions:
  1. Label one column "Seen in person" and the other "Seen on a screen"
  2. Every time a stranger behaves badly, put a mark in the relevant column
  3. Every time a stranger behaves decently (holding a door, returning change), put a mark in a second row of the same column
  4. At the end of the week, compare the ratios
  5. Write one sentence about which column your beliefs about "people today" had been drawing from
• Reflection questions:
  • How different were the two columns?
  • Which column did your gut treat as the truth about people?
• Frequency: Once, then repeat every six months

Exercise 2: The Honest Golden Age

• Objective: Restore the faded negatives to your memory of the "better" past
• Time required: 45 minutes
• Materials needed: Paper and pen; optionally a newspaper archive of the era
• Difficulty level: Intermediate
• Instructions:
  1. Write down the decade you think of as the time when people were more decent
  2. List ten things from that decade that were unkind, dishonest or cruel: how outsiders were treated, how children were disciplined
  3. If you lived through it, add three things you personally witnessed that you had not thought about in years
  4. If you did not, spend twenty minutes with a newspaper from that decade
  5. Rate, from 1 to 10, how confident you now are that the decade was more moral than the present
• Reflection questions:
  • Which items were hardest to recall, and why?
  • What did the "better" past look like for people unlike you?
• Frequency: Once per idealised era, revisited yearly

Exercise 3: The Decline Audit

• Objective: Test a specific decline belief against actual data
• Time required: 60 to 90 minutes
• Materials needed: Internet access; a notebook
• Difficulty level: Advanced
• Instructions:
  1. State one decline belief precisely: "People are less honest than in 1990"
  2. Decide, before looking, what evidence would count: trust surveys, lost-wallet experiments, volunteering rates
  3. Find the best longitudinal measure for that evidence and note what it shows
  4. Record whether the data support, contradict or fail to speak to your belief
  5. If you found nothing, write down what the absence of evidence should do to your confidence
• Reflection questions:
  • Did you want to reject the data when it disagreed with you?
  • Would you apply the same standard of proof to a belief that people had improved?
• Frequency: Once per major decline belief

Daily Practice

The Strangers' Scorecard

At the end of each day, recall every interaction you had with a stranger and sort it into "treated me well," "neutral" and "treated me badly."

• Suggested duration: 2 to 3 minutes
• Best time of day: Evening, while the day's interactions are still retrievable
• How to track progress: Keep a running tally for a month; most people find the "badly" column nearly empty

Weekly Challenge

Read the Past Straight

Once a week, read one primary source from the era you consider more decent: a newspaper front page, a court report, a diary. Note what it reveals about how people actually treated each other.

• Expected outcomes after 4 weeks: A more textured picture of the idealised past; reduced confidence in the decline story
• Journaling prompts for reflection:
  • What did I find that I had never been told about this era?
  • Which of today's complaints already appeared in that source?

12. For Specific Audiences

For Leaders and Managers

• Check the story against the numbers: Before acting on a sense that integrity or work ethic is slipping, pull the conduct, retention and engagement data. Stable numbers mean the slide is in perception.
• Beware generational folklore: "New hires aren't what they used to be" has been said about every cohort since the 1950s. Treat it as a known bias.
• Protect trust-based systems: Surveillance introduced to counter an imagined decline in honesty damages the honest majority most.

For Parents and Educators

• Age-appropriate explanation: "Grown-ups often think kids were nicer when they were young. That's mostly because they forgot the mean parts and now see every bad thing on their phones. Kids today are about as kind as kids have always been."
• Resist the "kids these days" reflex: Protzko and Schooler found that adults judge today's youth most harshly on the traits those adults excel at.
• Give children real history: A classroom that presents the past with its cruelty intact inoculates against the illusion better than a sanitised version.

For Healthcare Professionals

• Audit your narrative about patients: If you believe patients have become more demanding, check complaint and adherence rates in your own service first.
• Separate burnout from decline: Exhaustion lowers tolerance for ordinary friction, and the lowered tolerance can feel like a change in the people you serve rather than in you.
• Recognise the belief as a risk factor: A patient convinced the social world is deteriorating may be isolated, anxious and distrustful of care.

For Financial Professionals

• Do not price in a moral collapse that has not happened: Trust, as measured, has been stable for decades. Strategies built on a presumed rise in dishonesty carry a cost with no corresponding risk.
• Watch for the illusion in clients: "You can't trust anyone anymore" often precedes decisions to hoard cash or abandon diversified holdings.
• Check your own hiring: Discounting younger candidates on integrity grounds is a common expression of this bias in finance.

13. Interactions with Other Biases

Biases That Amplify This One

Biases That Counteract This One

Common Bias Chains

Negativity Bias → Availability Heuristic → Illusion of Moral Decline → Pessimism Bias → Withdrawal from cooperation

Outrage-weighted news intake (Negativity Bias) makes examples of wrongdoing easy to recall (Availability Heuristic), which produces the conviction that people are getting worse (Illusion of Moral Decline). That conviction generalises into expecting bad outcomes from others (Pessimism Bias), and a person who expects strangers to cheat stops cooperating with them.

Interrupting the chain: The cheapest intervention is the information diet, which starves the first link. The second is the personal-world check.

14. Cultural Perspectives

The belief appears to be close to universal. In Mastroianni and Gilbert's international archive, a majority reported decline on 86 percent of items across 59 nations, and in the 40 countries surveyed by Pew in 2002 and 2006, a majority in every one called moral decline at least a moderately big problem. The experimental and replication work, however, used American participants, so the mechanism has been tested in one culture while the belief has been measured in many.

The shape of the complaint varies. Cultures that venerate ancestors have a ready framework in which the past is morally authoritative. Cultures with a strong progress narrative hold the decline belief alongside faith in material improvement. Societies that have recently undergone war or rapid urbanisation may have genuine reasons to perceive change, which the illusion then exaggerates.

15. Myths and Misconceptions

16. Expert Insights

Adam Mastroianni and Daniel Gilbert found that people in dozens of countries believe morality has been declining for decades, although measures of current moral behavior show no such drop. They trace the illusion to two biases: people hear more about bad behavior now, and they remember the bad parts of the past less well. — Summary of Mastroianni & Gilbert, "The Illusion of Moral Decline," Nature 618 (2023), 782-789 https://doi.org/10.1038/s41586-023-06137-x

"Our fathers' age, worse than our grandfathers', has borne us, more worthless still, soon to produce offspring more vicious yet." — Horace, Odes 3.6.46-48 (23 BC); a literal English rendering of the Latin "aetas parentum, peior avis, tulit nos nequiores, mox daturos progeniem vitiosiorem"

People who excel at a particular trait are the ones most likely to believe that today's children are deficient in it, partly because they misremember their own childhood selves as already possessing it. — John Protzko and Jonathan Schooler, Science Advances, 2019 (paraphrasing)

17. Key Takeaways

1. The belief is nearly universal: Across 60 nations and 70 years of surveys involving roughly 12.5 million people, majorities said morality was declining on 84 to 86 percent of items.

2. The behaviour it describes has not changed: Surveys asking how strangers actually treat people, answered by about 4.5 million respondents between 1965 and 2020, show no trend.

3. The decline starts when you do: People perceive no decline before their birth and steady decline after it. The trend tracks the observer, not the world.

4. The people you know are exempt: Participants rated strangers as having declined and the people in their own lives as having slightly improved.

5. Two ordinary biases produce it: Negativity-skewed exposure to information about strangers, plus faster fading of negative memories. Mastroianni and Gilbert call this BEAM and present it as the leading candidate, not a proven cause.

6. The evidence is strong but young: The archival data are vast and an independent replication reproduced the key study, but the experiments are American and the mechanism is partly inferred.

7. Keep the standards, drop the history: Wanting people to be honest and kind does not require believing they once were more so.

18. Further Resources

Academic Papers

• Mastroianni, A. M., & Gilbert, D. T. (2023). The illusion of moral decline. Nature, 618(7966), 782-789. https://doi.org/10.1038/s41586-023-06137-x
• Protzko, J., & Schooler, J. W. (2019). Kids these days: Why the youth of today seem lacking. Science Advances, 5(10), eaav5916. https://doi.org/10.1126/sciadv.aav5916
• Eibach, R. P., Libby, L. K., & Gilovich, T. D. (2003). When change in the self is mistaken for change in the world. Journal of Personality and Social Psychology, 84(5), 917-931. https://doi.org/10.1037/0022-3514.84.5.917
• Walker, W. R., Skowronski, J. J., & Thompson, C. P. (2003). Life is pleasant, and memory helps to keep it that way! Review of General Psychology, 7(2), 203-210. https://doi.org/10.1037/1089-2680.7.2.203

Books

• Pinker, S. (2011). The Better Angels of Our Nature: Why Violence Has Declined. Viking.
• Rosling, H., Rosling, O., & Rosling Rönnlund, A. (2018). Factfulness: Ten Reasons We're Wrong About the World, and Why Things Are Better Than You Think. Flatiron Books.
• Gilbert, D. T. (2006). Stumbling on Happiness. Alfred A. Knopf.

Book Chapters

• Eibach, R. P., & Libby, L. K. (2009). Ideology of the good old days: Exaggerated perceptions of moral decline and conservative politics. In J. T. Jost, A. C. Kay, & H. Thorisdottir (Eds.), Social and psychological bases of ideology and system justification (pp. 402-423). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195320916.003.016

19. Summary Card

20. Glossary of Terms Used

21. Discussion Questions

For book clubs, classrooms, or self-reflection:

1. Think of the decade you consider the most decent in living memory. What was happening then to people unlike you: outsiders, the poor, children, minorities?

2. People see decline among strangers but slight improvement among the people they know. Why do we trust the stranger-based estimate over the personal one?

3. If the belief in moral decline is an illusion, should we try to correct it in public life, or does the complaint do useful work by keeping standards in view?

4. Adults judge today's youth most harshly on the traits those adults excel at. Where in your own "kids these days" complaints can you see your strengths doing the judging?

5. The experiments were run in the United States, and the mechanism is proposed rather than proven. What evidence would convince you the account is right, and what would convince you it is wrong?

最後更新：4 11月 2026