In 2006, a Science paper showed that subtly reminding people of money made them less helpful and more self-sufficient. The effect was replicated through the late 2000s, cited thousands of times, and built a research program. Then high-powered preregistered replications and a comprehensive meta-analysis found the effect was essentially zero. This is the kill shot that came in slowly.

The original demonstration was unforgettable.

Subjects walked into a lab and completed what looked like an unrelated language task. Some of them unscrambled sentences containing words like “salary,” “cost,” and “paid.” Others unscrambled sentences with neutral words. Then, on the way out, the experimenter “accidentally” dropped a box of pencils on the floor.

The participants who had been subliminally exposed to money words picked up fewer of the pencils than the participants who hadn’t. Same procedure, same experimenter, same dropped pencils. The only difference was a few words on a worksheet they’d already turned in. And it apparently changed how helpful they were.

Variations of the procedure piled up. A computer screensaver showing floating currency symbols (versus floating fish) made people put more physical distance between themselves and a stranger. A stack of Monopoly money on the desk made people work longer on a hard puzzle before asking for help, and donate less to a university student fund on the way out. The pattern that emerged was striking: subtle, unconscious reminders of money made people more self-sufficient and less prosocial. They asked for less help. They offered less help. They preferred to work alone, play alone, and sit further away from new acquaintances.

The 2006 paper that established this finding, published in Science by Kathleen Vohs, Nicole Mead, and Miranda Goode, became one of the most-cited demonstrations of social priming in modern psychology. It got picked up everywhere --- popular books, TED talks, corporate training programs, behavioral-economics workshops. The “money cues make people selfish” finding became a load-bearing claim in dozens of marketing and consulting frameworks.

Then, slowly, over the next decade, the empirical foundation gave way. High-powered preregistered replications failed to reproduce the effect. Meta-analyses with bias correction found the true effect was indistinguishable from zero. By 2019, the comprehensive meta-analytic verdict was in: when you look only at studies that controlled for publication bias and questionable research practices, money priming is essentially nothing.

This article walks through the original 2006 demonstrations, why they were so persuasive at the time, the replication failures that emerged from 2015 onward, the meta-analytic reckoning, and what an honest account of “money’s effect on behavior” looks like now.

The 2006 Vohs Original

The founding paper is Vohs, K. D., Mead, N. L., & Goode, M. R. (2006). “The Psychological Consequences of Money,” Science, 314(5802), 1154-1156. It contained nine experiments. The unifying claim was that subtle reminders of money --- primes too peripheral for participants to notice consciously --- pushed people into what the authors called a “self-sufficient orientation.” Money-primed participants preferred to work alone, play alone, sit further from others, and were less likely to either ask for or offer help.

The priming manipulations varied across experiments. Three of them are worth knowing specifically because they became the cultural touchstones:

Scrambled sentence task. Participants were given sets of jumbled words and asked to construct grammatical sentences. In the money-prime condition, the sets contained money-related words (“paid,” “salary,” “high-paying”); in the control condition, neutral words. After completing the task, dependent measures were collected. In one variant, the experimenter “accidentally” dropped a box of pencils; helpfulness was measured by how many pencils the participant picked up. In another, the participant was given an opportunity to donate to a student fund on their way out.

Monopoly money on the desk. In another paradigm, participants played a brief game of Monopoly and then either had a large amount of play money left in front of them, a small amount, or none. They then attempted a difficult puzzle that was actually unsolvable. The dependent measure was how long they would work on it before requesting help.

Currency screensaver. Participants worked at a computer that displayed either a screensaver of floating currency or one of floating fish underwater. They were then asked to move two chairs together for a face-to-face conversation with a stranger. The dependent measure was the physical distance between the two chairs.

Across the nine experiments, the results consistently pointed in the same direction. Money-primed participants picked up fewer pencils, donated less, worked longer before asking for help, sat further from strangers, and chose individual over group activities. The aggregate story was tidy: money --- even when subliminally evoked --- was making people behave more independently and less prosocially.

The paper was published in Science, one of the highest-prestige outlets in any field. The mechanism was theoretically appealing: it slotted into a growing literature on unconscious behavioral priming that included John Bargh’s elderly-walking study, Bargh’s warm-coffee study, the cleanliness-and-moral-judgment work, and many others. Within months, the money priming finding was being cited by economists, marketers, and behavioral scientists as evidence that everyday environmental cues --- what’s on someone’s screensaver, what’s printed on their coffee cup, what posters hang in a waiting room --- could meaningfully shape downstream behavior.

Why It Looked Convincing

The 2006 paper had several features that made it look bulletproof to the field at the time.

Nine experiments converging on the same story. A single experiment showing a counterintuitive effect can always be dismissed as chance. Nine experiments, with multiple priming paradigms and multiple dependent measures, all pointing in the same direction, looks much harder to dismiss. This is exactly the pattern that --- under classical assumptions about how psychology research is conducted --- should be very strong evidence for an effect. (Under more realistic assumptions about publication bias, file-drawer effects, and researcher degrees of freedom, the picture is different. But that more sophisticated reading of the literature was not yet standard in 2006.)

Independent replication in the late 2000s. Several follow-up studies, including some from independent labs, reported successful money-priming effects in adjacent paradigms. By 2010, the literature contained dozens of published studies all reporting positive money-priming results across countries and dependent measures. To anyone surveying the literature, the effect looked extraordinarily well-supported.

Theoretical resonance with the broader unconscious-priming program. Money priming arrived during the peak years of social-priming research. John Bargh’s lab and many others were producing a steady stream of findings showing that subtle environmental cues could change behavior. The money-priming finding fit perfectly into this framework. Theoretical coherence with adjacent literatures made the specific finding feel more credible than it would have looked in isolation.

A high-prestige publication venue. Science publication signals editorial confidence in a paper’s importance and methodological soundness. Vohs et al. 2006 was reviewed and accepted by one of the two most selective general-science journals in the world. For many readers, that was sufficient.

Vivid, communicable findings. Like the elderly-walking effect, the money-priming results were unforgettable in a single telling. “Show people money symbols on a screensaver and they sit further from a stranger.” “Have them unscramble sentences with money words and they pick up fewer dropped pencils.” The findings were perfect material for popular science writing, business books, and consulting decks. Cultural amplification compounded academic citation, and the framework grew.

By the early 2010s, money priming was a load-bearing concept in pop-behavioral-economics. It featured in Daniel Kahneman’s Thinking, Fast and Slow (2011) as one of the demonstrations of how the System 1 unconscious shapes behavior. It appeared in dozens of marketing and consulting frameworks. The empirical foundation looked solid; the cultural reach was enormous.

Then, beginning in 2015, the replication picture began shifting in earnest.

The Replication Storm Begins (Rohrer 2015)

The first major systematic challenge came from Doug Rohrer, Hal Pashler, and Christine Harris in 2015, who published “Do Subtle Reminders of Money Change People’s Political Views?” in the Journal of Experimental Psychology: General.

The Rohrer team conducted nine high-powered replication experiments, with a combined sample of over 1,800 participants --- a much larger total sample than the original Vohs work and most of the follow-ups. The targets were four money-priming effects from the closely related Caruso, Vohs, Baxter, and Waytz 2013 paper, which had reported that incidental exposure to money cues made people more supportive of inequality, free-market economics, group-based discrimination, and socioeconomic stratification.

The Rohrer replications were preregistered. The procedures were close to verbatim copies of the original methods. The sample sizes were large enough to detect even small effects with high statistical power.

The result: across all nine experiments, the weighted Cohen’s d was approximately 0.03 --- indistinguishable from zero. The four original effects from Caruso 2013 simply did not appear. Rohrer and colleagues also learned, after the fact, that the original Caruso team had run several follow-up studies that had also been null and were not reported in the original article --- a classic file-drawer pattern.

This was a serious blow. Rohrer 2015 was not just one failed replication of one finding. It was nine well-powered preregistered experiments targeting a closely related set of money-priming claims, all converging on the same null result. Conducted by a team that included Hal Pashler, who had spent years systematically replicating priming effects (the same group had previously published failed replications of professor priming, high-performance-goal priming, and other priming paradigms --- see our discussion in the Bargh elderly priming article).

The Rohrer paper triggered a debate that would crystallize the money-priming controversy.

The Meta-Analytic Reckoning (Vadillo 2016, Lodder 2019)

In 2015, in the same issue of Journal of Experimental Psychology: General that published the Rohrer replication failure, Kathleen Vohs published a reply: “Money priming can change people’s thoughts, feelings, motivations, and behaviors: An update on 10 years of experiments.” Vohs argued that the broader money-priming literature, by then encompassing 165 studies across 18 countries, was largely supportive of the original finding. She used what’s called a vote-counting approach --- comparing the count of successful replications to the count of failed ones --- to argue that the bulk of evidence still favored a real money-priming effect, and that the Rohrer failures should be interpreted as one set of null results within a much larger body of mostly positive evidence.

The response came from Miguel Vadillo, Tom Hardwicke, and David Shanks in 2016, in a paper titled “Selection Bias, Vote Counting, and Money-Priming Effects: A Comment on Rohrer, Pashler, and Harris (2015) and Vohs (2015),” also in Journal of Experimental Psychology: General.

Vadillo and colleagues made a methodological argument that has since become standard in replication-crisis analysis. Vote counting --- comparing how many published studies showed the effect versus how many didn’t --- is a deeply unreliable way to assess whether an effect is real, because it ignores publication bias. If the literature systematically publishes positive results and suppresses negative ones, then vote counting will inevitably favor “the effect is real” regardless of what’s actually happening in the underlying experiments.

The Vadillo team then applied meta-analytic bias-detection tools to the Vohs 2006 data and the broader money-priming literature. The headline finding: funnel plot analysis showed strong asymmetry in the Vohs et al. 2006 studies and the Caruso et al. 2013 studies --- exactly the pattern produced by selective reporting. The funnel plot asymmetry for Vohs’s original studies was statistically significant (t(12) = 5.42, p < .001). The Rohrer replications, by contrast, showed no such asymmetry --- consistent with the reasonable interpretation that they were unbiased samples from a true distribution centered near zero.

Beyond the funnel plot evidence, p-curve analysis of the money-priming literature estimated the average statistical power of the underlying experiments at approximately 0.18 in the main analysis and 0.25 in robustness checks. To put that in perspective: a literature with average power below 0.25 should produce a positive result less than one quarter of the time, but the published literature was reporting positive results far more often than that. The most economical explanation is that many of the published positive results are false positives produced by publication bias and questionable research practices.

The most comprehensive meta-analytic verdict came from Paul Lodder, How Hwee Ong, Raoul Grasman, and Jelte Wicherts in 2019, in a paper titled “A Comprehensive Meta-Analysis of Money Priming” in Journal of Experimental Psychology: General.

The Lodder team aggregated 246 money-priming studies. The naive overall effect size was g = 0.31 --- modest but apparently meaningful. But when they restricted the analysis to the 47 preregistered studies (the subset where publication bias and researcher degrees of freedom were most tightly controlled), the average effect size collapsed to g = 0.01. Not “smaller than the original literature suggested” --- essentially zero. The pattern is consistent with what you’d expect if the published positive findings were largely the product of selective reporting on underpowered studies, and the true effect, in carefully designed experiments, was negligible.

The 2019 meta-analysis is one of the cleanest empirical demonstrations in the social priming canon. Same construct, two analytical lenses: when you include the full noisy biased literature, you see a modest effect; when you restrict to preregistered studies, the effect disappears. The difference between those two estimates is, in effect, the size of the publication-bias and questionable-research-practices distortion that was holding the original literature up.

Vohs’s Defense And Why It Falls Short

Kathleen Vohs has continued to defend money priming. Her position has evolved over time, but the core arguments fall into a few categories.

The heterogeneity defense. Vohs argues that money priming is moderated by many variables --- the meaning of the prime, the ambiguity of the situation, the participant’s prior beliefs, the cultural context --- and that average effect sizes across a heterogeneous literature mask real effects in specific subgroups. There’s a kernel of truth here: psychological effects often are moderated by context. But this defense becomes problematic when, as in the money-priming case, the well-controlled preregistered subset (which would be expected to detect any robust effect, even after accounting for moderators) shows essentially zero average effect. If the moderators were doing the work that the heterogeneity argument requires, you would still expect to see some clear positive subset in the preregistered literature. Lodder 2019 did not find such a subset.

The procedural-fidelity defense. Vohs and others have argued that failed replications deviated subtly from the original procedures in ways that mattered. This is a recurring response across the priming literature (see also Bargh’s reply to Doyen on elderly priming). It’s also a hypothesis that becomes increasingly hard to sustain as more replication teams try more variations and continue to find null results. By the time you’ve explained nine null replications by procedural deviations, the “true” procedure that produces the effect is getting suspiciously narrow.

The vote-counting defense. Vohs 2015 argued that 165 studies across 18 countries showing positive money-priming effects can’t all be wrong. As Vadillo 2016 demonstrated rigorously, this argument fails when the underlying literature shows clear signs of publication bias. The count of positive results in a biased literature tells you about the bias, not about the underlying effect.

The field has, by and large, moved on. The money-priming concept is no longer a load-bearing piece of the active social-psychology research program in the way it was in 2010. Younger researchers entering the field do not generally treat the original Vohs findings as a foundation to build on. The Lodder 2019 meta-analysis is the verdict that most working researchers now cite when the topic comes up.

This isn’t unanimous. Vohs has continued to defend the original findings, and a few researchers continue to publish in the broader money-priming paradigm. But the institutional center of gravity has shifted decisively. The honest summary of where the empirical literature now stands: the subtle, unconscious “money cues change behavior” version of the claim is not supported by the best-controlled evidence.

What This Means Alongside The Broader Social-Priming Crisis

Money priming is not an isolated case. It sits inside a broader pattern of canonical social-priming effects from the 1996-2010 era that have failed rigorous replication. The list is long and worth understanding as a unit.

Bargh elderly priming (1996). The iconic finding that priming participants with elderly-stereotype words made them walk more slowly down a hallway. Failed replications beginning with Doyen 2012 cast serious doubt on the effect; see our detailed treatment.

Professor priming. Priming with words related to professors supposedly improved performance on knowledge tests. Multiple failed replications, including Shanks et al. 2013.

High-performance-goal priming. Subliminal exposure to achievement-related cues supposedly improved subsequent task performance. Failed replication by Harris, Coburn, Rohrer & Pashler 2013.

Warm-coffee priming (Williams & Bargh 2008). Holding a warm cup of coffee supposedly made participants judge a stranger as warmer. Failed replication.

Cleanliness-and-moral-judgment (Schnall et al.). Activating cleanliness concepts supposedly made participants judge moral transgressions less harshly. Mixed replication record.

Money priming (Vohs et al. 2006). The subject of this article. Failed replications (Rohrer 2015, others) and decisive meta-analysis (Lodder 2019) showing essentially zero effect in preregistered designs.

The 2012 Kahneman open letter --- the famous “I see a train wreck looming” warning to the social-priming research community --- was responding to exactly this accumulating pattern. By 2012, the failures in the elderly-walking line were already visible. By 2019, the money-priming line had joined them in the same status: an effect that had been culturally enormous in its day, that no longer survives rigorous testing.

The pattern matters because it points at a class of claim that should now be approached with skepticism. “Subtle environmental cue X produces dramatic behavioral effect Y through unconscious processing” is, as a category, much weaker than the cultural literature suggests. Not every such claim is false --- semantic and evaluative priming (the cognitive forms) are well-supported. But the behavioral version --- exposure to a stimulus producing a measurable downstream change in real-world action --- is a category where the prior should be skeptical, especially when the supporting evidence comes primarily from small-sample experiments published before preregistration norms took hold.

What’s Honest To Say About Money’s Effect On Behavior Now

It’s important to be precise here, because “money priming doesn’t replicate” is not the same as “money doesn’t affect behavior.”

Money obviously affects behavior in conscious, situational, consequential ways. Economists have studied this exhaustively. People work harder for higher pay (above a certain range, with complications). People allocate time differently when they have more or less money. People make different consumption choices when prices change. People take different jobs when wages differ. The full apparatus of microeconomics rests on the demonstrated fact that money --- when consciously perceived and consequentially attached to outcomes --- shapes behavior in predictable ways. None of this is in dispute.

The subliminal, subtle-priming version of the claim is different and much weaker. What Vohs 2006 claimed wasn’t that money affects behavior. It was that barely perceptible reminders of money --- a few words in a sentence-unscrambling task, a screensaver running in the background, a small stack of play money on a desk --- could produce measurable downstream effects on prosocial behavior. This is a much narrower and more specific claim, and it is this narrower claim that the replication failures and meta-analyses target.

Conscious, salient money cues might still have some effects. It is plausible --- though far from established at the level once claimed --- that prominent reminders of money (a large salary number displayed on screen, an explicit conversation about wealth) can shift behavior in some contexts. The strong version of money priming as practiced in the original literature was about subtle cues operating through unconscious channels. Even after replication failures invalidate that specific version, more conscious framing effects probably remain partially supported. The honest summary is: the strong, subliminal, automatic-unconscious version of money priming is not supported; weaker, more conscious framing effects are a separate question, partially supported by other literatures (especially in economics and behavioral decision research).

Cultural cues about money matter through different mechanisms. If you grow up in an environment where wealth is central to identity, your behavior will probably differ in some ways from someone who grew up where wealth was peripheral. This is not the same claim as the one Vohs et al. were making. It’s a claim about long-run socialization rather than a claim about momentary unconscious priming. The longitudinal sociological evidence on socialization-around-money is much stronger than the experimental priming evidence.

For decisions that depend on whether subtle environmental cues about money meaningfully shape downstream behavior, the empirical foundation is now weak. For decisions that depend on whether explicit, consequential financial incentives shape behavior, the foundation is strong but it’s a different question being answered by different literatures.

What This Means For Strategists

Three takeaways for consultants, marketers, and leaders who have absorbed the money-priming framework into their thinking.

1. The “ambient money cue” school of marketing is empirically thin. A subset of behavioral-marketing consulting frameworks recommend placing money-related visual cues in environments where you want certain behaviors --- currency imagery in fundraising pitches, wealth-themed background music in luxury retail, dollar-sign motifs in pricing pages. The empirical justification for these recommendations typically traces back to the Vohs 2006 paper and the broader money-priming literature. After 2019, that empirical justification is weak. The honest reading is: ambient money cues probably do not meaningfully shift downstream behavior in the consistent, replicable ways the original literature claimed. If your marketing strategy depends on ambient money cues producing measurable downstream effects, you are betting on a paradigm that the empirical evidence no longer supports.

This doesn’t mean you have to abandon all marketing decisions tied to environmental design. It means the mechanism you should hypothesize is different. Conscious framing (an explicit price, a clear value proposition, a salient comparison) probably matters. Subtle ambient cues probably don’t, at the magnitudes the priming literature once claimed. Reallocate strategic attention from the latter to the former.

2. “Abundance vs scarcity priming” frameworks are mostly built on contested ground. Several pop-business frameworks instruct leaders to “prime abundance” or “prime scarcity” in their teams by manipulating environmental cues --- choice of vocabulary, framing of metrics, visual surroundings. The underlying psychological theory often draws on the same priming literature that includes money priming. The empirical support for these frameworks at the magnitudes they typically promise is much weaker than their packaging suggests. Treat such recommendations as speculative cultural framing, not as evidence-based interventions.

3. The pattern repeats --- be skeptical of any “subtle cue, dramatic behavioral effect” claim. The deeper lesson from the money-priming story is methodological. When you encounter any claim of the form “exposure to cue X produces dramatic behavioral effect Y through unconscious mechanisms,” your default prior should be skeptical. This is the class of claim that has performed worst in the replication crisis. Whether the specific finding is money priming, elderly priming, color priming, ambient-music priming, or any other “subtle cue, large effect” framing, the base rate of such claims surviving rigorous preregistered replication is much lower than the popular literature would suggest. Demand the preregistered evidence specifically. If only the un-preregistered evidence exists, treat the claim as a hypothesis, not a finding.

The Vohs 2006 paper was an unusually clean example of a research program that grew enormous on a foundation that, in hindsight, was not load-bearing. The pattern it follows --- vivid finding, prestigious publication, rapid academic citation, immediate translation into business consulting, then a slow accumulation of replication failures, then a decisive meta-analytic verdict --- is the modal trajectory for behavioral-science findings of this era. Recognize the pattern. Adjust priors accordingly.

Sources

This article is part of an ongoing series on famous behavioral-science studies that did not survive replication. Closely related: the Bargh elderly-priming article, which covers the parallel social-priming kill shot, and the broader context of the Kahneman “train wreck” letter. Other entries cover ego depletion, power posing, the growth mindset, the marshmallow test, the facial feedback hypothesis, and the Stanford Prison Experiment. The full hub lives at /replication-crisis/.

If you’ve built marketing, product, or strategy decisions on money-priming or broader social-priming claims and want a careful evidence review, book a consultation.

FAQ

Did Vohs and colleagues fabricate the original 2006 data? No. There has been no suggestion of fraud or fabrication in the Vohs 2006 paper or in any of the subsequent money-priming work from her lab. The question is not whether the data were faked. The question is whether the underlying effect, as it exists in well-controlled preregistered designs, is meaningfully different from zero. The Lodder 2019 meta-analysis says no.

How can the original 2006 paper and all the follow-up studies be wrong if so many of them showed positive results? This is the central question Vadillo et al. 2016 addressed. The answer is publication bias plus questionable research practices applied to underpowered studies. When the average statistical power of the underlying experiments is around 0.18-0.25, the literature should be reporting null results most of the time. But the published literature is reporting positive results frequently. The most economical explanation is that the null results are being suppressed (file drawer) and the positive results that do get published often involve flexibility in analytic choices that inflate false-positive rates. The aggregate picture is a literature that appears to show a robust effect but is mostly an artifact of selective reporting.

Has Vohs retracted any of the original work? No. Vohs has continued to defend the original findings and the broader money-priming research program. The 2015 reply to Rohrer is her most explicit defense. The field has largely moved on independently of any retraction.

Does this mean money has no psychological effects? No. Money obviously has psychological and behavioral effects when it is consciously perceived and consequentially attached to outcomes --- economists and decision researchers have established this extensively. What’s contested is the specific narrower claim that subtle, subliminal reminders of money --- words in a sentence-unscrambling task, a screensaver, play money on a desk --- produce measurable downstream effects on unrelated behavior through unconscious channels. That narrower claim is what failed replication.

What about the famous “money makes you mean” popular framing? The popular “money makes you mean” framing --- based partly on Vohs 2006 and partly on related work like Piff et al.’s wealth-and-unethical-behavior studies --- has weakened substantially as preregistered replications and meta-analyses have come in. The literature on wealth and prosocial behavior more broadly is more nuanced than the popular framing suggests. Some effects in some conditions exist; the strong universal “wealthy people are meaner” claim is not well-supported in the highest-quality evidence.

Should I trust Thinking, Fast and Slow’s chapter on priming? The dual-process framework of Thinking, Fast and Slow is not what’s contested. The specific empirical demonstrations in the priming chapter --- including the money-priming example --- should now be read with substantial skepticism. Kahneman himself acknowledged this implicitly in his 2012 open letter and has continued to publicly engage with the replication-crisis findings. The book is still valuable for its broader framework, but several of its specific examples (priming included) need updating in light of subsequent failed replications.

Does any aspect of money priming survive the meta-analytic verdict? A weak honest position: it remains possible that money priming produces small effects in some specific paradigms under some conditions, but the average effect in preregistered, well-controlled designs is essentially zero. If a true effect exists, it is much smaller and more conditional than the original 2006 paper and its follow-ups claimed. For practical purposes --- building marketing campaigns, structuring environments, designing experiments --- treating money priming as having no reliable behavioral effect is the appropriate prior.

What’s the broader lesson for consumers of behavioral-science research? Vivid findings published in prestigious outlets and amplified through popular books deserve no exemption from rigorous replication testing. The money-priming story shows how a research program can grow enormous on a foundation that turns out, when finally tested at scale, to be much weaker than its institutional success suggested. When you encounter the next “subtle cue, dramatic behavioral effect” claim, ask specifically for the preregistered replication evidence. If it doesn’t exist, the claim is a hypothesis, not a finding.

replication-crisis money-priming social-priming social-psychology evidence-evaluation

Free Tool

Built for Experimentation Teams

GrowthLayer is the experimentation platform I built for CRO teams --- test management, AI-powered insights, and pattern recognition across your entire program.

Explore GrowthLayer → (opens in new tab)

· Start Free →

Share this article

LinkedIn (opens in new tab) X / Twitter (opens in new tab)

Copy link

Go deeper

Methodology The PRISM Method Case Studies $30M+ in Results Work Together Services & Mentoring

Experimentation and growth leader. Builds AI-powered tools, runs conversion programs, and writes about economics, behavioral science, and shipping faster.

About LinkedIn Newsletter

← Previous

The Hawthorne Effect: The Famous Workplace Story That The Data Don’t Support

Next →

Broken Windows Theory: The Atlantic Essay That Reshaped Policing On Weak Evidence

replication-crisis money-priming social-priming social-psychology evidence-evaluation
Share this article
LinkedIn (opens in new tab) X / Twitter (opens in new tab)
Atticus Li

Experimentation and growth leader. CXL-certified CRO practitioner, Mindworx-certified behavioral economist (1 of ~1,000 worldwide). 200+ A/B tests across energy, SaaS, fintech, e-commerce, and marketplace verticals.

About LinkedIn Newsletter