False-consensus effect is the tendency to overestimate how widely one’s own beliefs, preferences, and behaviors are shared. Ross, Greene, and House documented it in 1977: people who held a given view consistently judged it more common than those who held the opposing view did, each projecting their own position onto the population. We use ourselves as the default sample, so what feels obvious to us feels obvious to “most people” — a comfortable illusion that rarely gets checked against real data.
Manipulation exploits this from both directions. It can hand you a claim that “everyone agrees” and let your own projection supply the evidence, since a stated majority that mirrors your bubble feels self-evidently true. Or it can manufacture the appearance of consensus outright — astroturfing, coordinated comment campaigns, and repetition that makes a fringe position feel widely held (a companion to the mere-exposure effect, T8.16). Either way the tactic is self-reinforcing: once dissent is framed as fringe, doubters fall silent, and their silence makes the false majority look larger still, a spiral of manufactured agreement.
The defense is to insist on a real sample. Ask where a “most people think” claim actually comes from — a poll, a dataset, a varied cross-section — and step deliberately outside your own circle to test it, because a consensus that only your in-group holds is no consensus at all. Distinguish many independent voices from a few amplified ones, and voice your own dissent when you have it: stating a contrary view breaks the illusion for everyone watching and reveals whether the majority was ever there.