R — Original Text (Reading)
In fact, these anti-drug ads not only failed to reduce drug use, but also backfired, increasing the frequency of drug use. After watching the ads, adolescents aged 12 to 18 actually showed an increased desire to smoke marijuana. Why did this happen? Because these ads made marijuana smoking more public. The industry association solemnly warned people online, "In the United States, only 37% of the music consumers play is paid for." If nothing unexpected happens, it is very likely to have the opposite effect. Only a little more than half of people are willing to pay for music? Wow. It seems that paying for music symbolizes that you are a big fool.
— Jonah Berger, Chapter 4 Public — Counterexample? The Pharmaceutical Industry
I — Methodological Skeleton (Interpretation)
The publicity backlash effect reveals a counterintuitive rule: openly discussing a bad behavior in order to prohibit it actually increases the incidence of that behavior.
Operating mechanism:
- Social proof principle— People decide what to do by observing "what others are doing." When publicity says "many people are doing X," the signal people receive is not "X is bad," but "X is very common."
- Visibility increases imitation— Publicity makes hidden behavior visible, and only visible behavior can be imitated. Anti-drug ads made drug use more visible, and instead provided a template for imitation.
- Normalization effect— Citing large-scale data on bad behavior ("37% of music is pirated") makes that behavior appear to be the "normal choice," and makes those who behave correctly feel like "fools" instead.
The correct approach: do not publicize the scale of the bad behavior, but rather publicize the prevalence of the correct behavior. Transform the message "do not do X" into the message "most people are doing the right thing."
A1 — Application in the Book (Past Application)
Case 1: Anti-drug ads increased drug use rates
- Problem: The U.S. government spent nearly 1 billion dollars from 1998 to 2004 on anti-drug publicity for adolescents, but the effect was counterproductive.
- Use of the methodology: Anti-drug ads showed scenes and frequencies of drug use, making hidden drug use public and visible. For adolescents, the message conveyed by the ads was not "drugs are harmful" but "many people are using drugs."
- Conclusion: Publicity is a double-edged sword — making bad behavior visible is equivalent to providing social proof and increases the probability of imitation.
- Result: Anti-drug ads increased the frequency of marijuana use among adolescents, and spending nearly 1 billion dollars backfired.
Case 2: Petrified Forest National Park petrified wood theft sign
- Problem: How to reduce the theft of petrified wood in the park?
- Use of the methodology: Psychologist Cialdini tested two slogans. The first emphasized "too many people steal petrified wood" (providing social proof for stealing), while the second changed to "to preserve ecological balance, please do not take petrified wood" (focusing on the correct behavior rather than the wrong behavior).
- Conclusion: A slogan that highlights bad behavior is equivalent to modeling that behavior; only a slogan that highlights the correct behavior can effectively reduce bad behavior.
- Results: The slogan highlighting the correct behavior successfully reduced the amount of petrified wood stolen, while the slogan highlighting the scale of theft actually increased theft.
A2 — Trigger Scenarios (Future Trigger) ★
In what situations will users need this skill?
- Users are designing "do not do X" publicity materials for nonprofit organizations or government departments (anti-drug, anti-bullying, anti-piracy).
- Users want to cite statistics on bad behavior to "warn" the public (such as "1 in 3 students experiences bullying").
- Users are promoting a compliance or safety culture within a company and are considering using "violation cases" to warn employees.
Language Signals (activate when these appear in the user's words)
- "We want to make an anti-drug/anti-bullying/anti-piracy campaign"
- "Data shows X% of people are doing (bad behavior), and we want more people to know"
- "How can we use data to make people realize the severity of the problem?"
- "Why is publicizing 'do not do X' not effective?"
- "Why did our public service announcement actually make the problem worse?"
Distinction from adjacent skills
- Difference from
visibility-design: Visibility design focuses on "making the correct behavior visible," while the boomerang effect of publicness focuses on "the danger of making the wrong behavior visible." The two are the positive and negative sides of the same principle—the former is a positive application, and the latter is a negative warning. - Difference from
emotion-arousal-model: The emotional arousal model focuses on the emotional attributes of content, while the boomerang effect of publicness focuses on the "behavioral modeling" attributes of content. Even if the emotion is right (high-arousal fear), if the content makes the bad behavior more visible, it will still backfire. - Difference from
social-proof: Social proof is the underlying psychological mechanism, while the boomerang effect of publicness is a misuse scenario of social proof.
E — Executable Steps (Execution)
After the skill is activated, the agent should execute the following steps:
-
Review whether the current content is "publicizing bad behavior"
- Check whether the publicity material cites the incidence, scale, or specific scenarios of the bad behavior.
- Completion standard: Mark all content elements that may provide "reverse social proof."
-
Shift the focus from "prohibiting bad behavior" to "promoting correct behavior"
- Replacement strategy: Change "X% of people use drugs" to "Y% of teenagers choose to stay away from drugs."
- Replacement strategy: Change "too many people steal petrified wood" to "the vast majority of tourists are protecting our natural heritage."
- Completion standard: The main body of the publicity content shifts from displaying bad behavior to displaying the prevalence of correct behavior.
- Stop condition: If the content itself is the promotion of positive behavior (and does not involve displaying bad behavior), then no adjustment is needed; skip to step 3.
-
Verify whether the revised content makes the correct behavior the visible majority
- Test: Does the revised content make the audience feel that "most people are doing the right thing"?
- Completion criteria: Confirm that the correct behavior is publicly visible and the undesirable behavior is invisible.
B — Boundary ★
Do not use this skill in the following situations
- When the undesirable behavior is indeed extremely rare and requires warning— If the incidence of the undesirable behavior is extremely low (such as school shootings), publicizing it is unlikely to provide meaningful social proof, and the warning effect may outweigh the risk of backlash.
- When the audience has independent judgment— Trained professionals are unlikely to be influenced by "social proof" to imitate undesirable behavior.
- When the information goal is "providing facts" rather than "changing behavior"— The purpose of news reporting is to inform about facts, not to change behavior, and the analytical framework of the backlash effect does not fully apply.
Failure modes the author warns about in the book
- Anti-drug ads increase drug use(ce-007): The U.S. government spent nearly 1 billion dollars on anti-drug ads, which instead increased the frequency of marijuana use among teenagers. The ads made drug use more public and provided social proof that "many people are using."
- Music piracy campaigns provide incorrect social proof(ce-014): "Only 37% of music is paid for" implies that not paying is the norm, making those who pay feel like "fools" instead.
- Petrified wood theft warnings increase theft(related to ce-007/v10): Posting "too many people steal petrified wood" is equivalent to demonstrating theft behavior, and instead increases theft.
The author's blind spots / limitations of the era
- The cases in the book mainly come from public health and environmental protection, and do not discuss similar backlash in the commercial field (such as ads saying "do not buy competitors" possibly instead giving competitors exposure).
- It does not discuss the problem of a "negative information vacuum"—if the scale of undesirable behavior is not shown, the public may underestimate the risk. In scenarios requiring urgent warnings (such as a pandemic), completely withholding data on undesirable behavior may also be inappropriate.
- It does not discuss the possibility of "segmented communication"—showing undesirable behavior data to adults may not produce backlash (because they already have fixed behavioral patterns), but it may backfire with teenagers.
Adjacent methodologies easily confused
- "Information transparency": Publicity backlash is not opposed to information transparency, but rather reminds that the way information is presented can trigger the social proof effect. The key is not "whether to disclose," but "what to disclose—the prevalence of undesirable behavior or the prevalence of correct behavior."
Related skills
- contrasts-with: visibility-design — Visibility design makes correct behavior visible, while publicity backlash reveals the danger of making wrong behavior visible; the two are the two sides of the same principle
Audit information
- Verification passed: V1 ✓ / V2 ✓ / V3 ✓
- Test pass rate: To be tested (see test-prompts.json)
- Distillation time: 2026-05-03