Your critical thinking gym
Home
The Commandments
Join the Club
Your critical thinking gym
● Cognitive bias
Named by economist Timur Kuran and legal scholar Cass Sunstein, 1999.
When a claim keeps showing up everywhere, it starts to feel more true and more urgent, not because new evidence turned up, but because it kept getting repeated.
Built from two linked mechanisms: informational cascade and reputational cascade. Not the same as the conformity/bandwagon effect or the illusory truth effect (Section 01).
A self-reinforcing process of collective belief formation in which an initial claim or event triggers a chain reaction of media coverage and public discussion, so the claim's perceived plausibility, importance, or urgency keeps rising because it keeps appearing, not because new evidence is arriving to support it.
Plain version
Something gets repeated so much, by so many people, that it starts to feel obviously true and urgently important, even though nobody's necessarily checked anything new lately, and the "everyone" repeating it might all be echoing the same original source.
The two pieces it's built from
Informational cascade: a person partly updates their own belief just by noticing what others already seem to believe, on the logic that if this many people are saying it, there's probably something to it. Reputational cascade: a person then repeats or publicly endorses the claim to stay in step with the apparent consensus, sometimes even holding private doubts, because visibly disagreeing carries a social cost. Kuran and Sunstein's model stacks these two together on top of the older availability heuristic (Tversky and Kahneman), using how easily something comes to mind as a stand-in for how true or common it is.
Not the same as conformity/bandwagon effect
The conformity/bandwagon effect is a person's general tendency to align belief or behavior with a perceived majority, in any given moment. An availability cascade is the larger, sequential, media-driven process by which that "perceived majority" itself gets built up or inflated over time through repetition. Bandwagon is one gear inside the cascade; it isn't the whole machine.
Not the same as illusory truth effect
The illusory truth effect is about one person hearing the exact same statement repeated and rating it as more true each time; it works even with no social signal at all, alone in a lab. An availability cascade layers social and reputational dynamics on top of that individual effect, at the scale of public discourse, with real-world actors amplifying the repetition on purpose.
Scope
Gets confused with
Conformity/bandwagon effect and illusory truth effect: precise distinctions above and in Section 09.
Deliberately one of the lowest scores on the whole site. A claim being "everywhere" doesn't feel like social pressure from the inside, it feels like a well-corroborated fact. There's an always-available, reasonable-sounding cover story (it's been reported everywhere, someone must have checked), and being alarmed or informed alongside everyone else feels good, like vigilance rather than vulnerability. Nothing here trips a wire.
Why your brain does this
In a small group, "how easily does this come to mind" was a reasonable stand-in for "how common or serious is this." If you'd heard about a danger repeatedly from people around you, word had usually spread because it was genuinely common or serious, direct testimony from your own community. Trusting the apparent judgment of the group was also a cheap, usually reliable way to form accurate beliefs without personally verifying everything, and visibly going along with what the group believed protected your standing in a community where reputation was a real survival stake. Neither shortcut is broken. Both are fast.
Where it misfires
Mass media and social platforms break the assumptions both shortcuts depend on. Coverage frequency no longer tracks real-world frequency: one alarming story can be re-reported, reposted, and recycled across dozens of outlets, so it looks like independent corroboration from many separate sources when it's really one claim echoing itself. And "the apparent judgment of the group" can be manufactured rather than genuinely aggregated (Section 03), so the exact signal both shortcuts rely on can be faked at scale in a way no ancestral environment ever had to guard against.
The live academic debate, briefly
The traditional Tversky and Kahneman heuristics-and-biases framing would treat this as a specific error, the availability heuristic operating at social scale. Gerd Gigerenzer's ecological-rationality research would instead treat both pieces here as well-suited strategies for real environments with limited time and information, even though they fail badly once the environment itself changes (echo chambers, algorithmic reposting) in ways evolution never had to account for. A related note specific to the reputational half: Hugo Mercier and Dan Sperber's argumentative theory of reasoning argues that reasoning didn't evolve mainly to find truth alone, but partly to justify decisions and win arguments with other people, which is exactly what a social, reputation-tracking animal would do when it repeats a claim to stay in step with perceived consensus. This page doesn't take a side between these framings; all are live in the research.
How it activates, step by step
1. A claim, event, or risk enters public discourse: a news story, a viral post, a dramatic anecdote.
2. Some people update their belief simply because others already seem to believe it (informational cascade), the availability heuristic operating at social scale.
3. Others repeat or publicly endorse the claim to stay in step with the perceived consensus, sometimes despite private doubt (reputational cascade), which increases how often the claim actually appears.
4. Rising repetition and apparent unanimity make the claim easier to recall and feel more obviously true, compounding with the illusory truth effect, often outrunning what the evidence at hand actually supports.
5. Public concern or policy response scales to the perceived size of the issue, which is not always the same as its actual size.
An availability cascade doesn't need a deliberate architect to get started, but several propaganda techniques exist specifically to start or accelerate one on purpose.
Fakes the exact signal an informational cascade depends on: the appearance that "everyone" already believes or supports a claim. Once that appearance exists, real people's own informational-cascade shortcut treats it as genuine social proof and starts the cascade for real.
Plants seemingly independent voices repeating the same claim. What looks like many people's separate corroboration is actually one coordinated source multiplied, directly feeding the illusion that a cascade needs to look organic.
Ad nauseam (repetition)
Raises a claim's raw availability through sheer repetition. This is the mechanical fuel a cascade runs on once it's underway, and it's what compounds with the illusory truth effect at step four of the activation sequence in Section 02.
Floods the environment so no one person can realistically verify every claim individually. That makes people more likely to fall back on "well, it's everywhere, someone must have checked," extending an availability entrepreneur's cover.
Outside politics
Marketing and PR use manufactured buzz constantly and non-maliciously in its mildest form: a wave of paid influencer posts timed to a product launch makes it look like something is "already the thing everyone's talking about," triggering the same informational-cascade shortcut a genuine grassroots trend would. In its more predatory form, pump-and-dump investment schemes deliberately flood forums and group chats with the same "insider tip" from many apparently unconnected accounts, engineering a reputational cascade among people afraid of being the one who missed it.
What increases it
Strong: topic-relevant prior anxiety or unfamiliarity. Health scares, new technologies, and unfamiliar risks raise baseline uncertainty, which increases reliance on social cues over personal verification.
Strong: identity or group stakes attached to the claim. A cascade tied to in-group standing (see In-Group Bias) gains reputational force faster, since repeating it signals loyalty as well as belief.
Mechanical: high-repetition, algorithmically amplified media environments. The same person is measurably more exposed to cascade-style repetition on a platform built to resurface popular claims than on one that isn't.
Moderate: time pressure or crisis framing. A shortened decision window pushes people toward the fast informational shortcut instead of independent verification.
What does NOT predict it
This site doesn't have a strong, specific citation testing whether general intelligence or education level predicts resistance to availability cascades in particular. Most of the available research targets individual biases like confirmation bias or the availability heuristic on their own, not the full social-cascade dynamic. We're flagging that gap rather than guessing at an answer we can't back up.
State-dependent factors
Yes, heavily so. Kuran and Sunstein's own account of how cascades take off leans on strong emotional arousal: fear and dramatic framing pull attention toward the vivid, available claim and away from slower verification, which is part of why "availability entrepreneurs" reach for alarming framing rather than measured framing in the first place.
The pattern across all three: in each case, the number of actual firsthand incidents behind the story was small relative to the scale of public reaction. What scaled fastest was the coverage and repetition, not the underlying incident count, which doesn't automatically mean the original concern was baseless (Alar's case genuinely is disputed), only that the pace of public alarm reflected media amplification dynamics as much as it reflected new evidence.
First named
Economist Timur Kuran and legal scholar Cass Sunstein coined "availability cascade" in "Availability Cascades and Risk Regulation", Stanford Law Review, 51(4), 683 (1999).
The landmark case analysis
Rather than a single lab experiment, Kuran and Sunstein built the concept around three real-world case studies: the Love Canal toxic-waste scare, the 1989 Alar pesticide scare, and the 1996 TWA Flight 800 crash (initially misattributed to terrorism in extensive media speculation before investigators found a mechanical cause). In each, they argued, spreading public panic led to escalating, expensive government action against risks that later analysis judged to have been substantially exaggerated relative to the evidence available at the time.
Direct lineage from the availability heuristic
The cascade model builds explicitly on the availability heuristic, the cognitive shortcut first described by Amos Tversky and Daniel Kahneman in "Availability: A Heuristic for Judging Frequency and Probability" (Cognitive Psychology, 5(2), 1973), the finding that people judge how common or likely something is by how easily examples come to mind. Kuran and Sunstein's contribution was showing how that individual-level shortcut compounds into a public, self-reinforcing process once media amplification and social reputational pressure get layered on top.
Since then
Daniel Kahneman later folded the concept into Thinking, Fast and Slow (2011), citing celebrity divorces and airplane crashes as everyday cases of media over-coverage inflating perceived frequency. Sunstein extended the mechanism further in On Rumors (2009), applying cascade dynamics specifically to how misinformation spreads.
Cross-tradition note
See Section 02 for how Gerd Gigerenzer's ecological-rationality framing and Hugo Mercier and Dan Sperber's argumentative theory of reasoning each add a genuinely distinct angle here, Gigerenzer on why both underlying shortcuts are ecologically reasonable, Mercier and Sperber on why the reputational-cascade half in particular fits a reasoning-for-social-justification account.
Score: 1 out of 5, deliberately one of the lowest on the whole site (see the breakdown in the card after Section 01).
The in-the-moment question
Before reacting to a claim that feels urgent: am I believing this because I've personally seen new evidence, or just because I keep seeing the same claim repeated? Try to name where you actually first encountered it, and whether the versions you've since seen trace back to that one source or to something genuinely independent.
Why this is so hard to catch
Repetition itself masquerades as independent confirmation. From the inside, "many people independently checked this and it's true" and "many people are just forwarding the same post" feel exactly the same; there's no internal marker distinguishing genuine corroboration from an echo. That's what makes this one of the hardest biases on the site to catch while it's happening.
Debiasing research note
Kuran and Sunstein's institutional proposals are reasoned recommendations, not interventions that have been experimentally tested against real-world cascades, which is a genuinely hard thing to run a controlled trial on. This is one of the areas on this site where the counter-moves rest more on theory than on measured results, and that's worth saying plainly rather than overselling.
Commonly co-occurs with
Often mistaken for
A person's general tendency to align with a perceived majority at any given moment; it's one gear inside a cascade, not the same as the whole sequential, media-amplified process. See Section 01.
The individual-level finding that repeated exposure to the same statement increases perceived truth, even with zero social signal. An availability cascade adds the social and reputational layer on top of that at public scale. See Section 01.
Compounding effect
The availability heuristic is the individual-level engine that makes "easy to recall" feel like "common or true." Conformity/bandwagon effect supplies the social-conformity fuel once a claim looks widely held. The illusory truth effect is what makes the repeated claim feel true on each new encounter. An availability cascade is essentially all three compounding together over time at the scale of public discourse, and it compounds further once identity or group stakes attach to the claim (see In-Group Bias, Section 04).
Primary sources
Accessible reading