Separate "many people worry about X" from "X is actually likely"

The fact that a risk is widely discussed is evidence about social dynamics, not about probability.

Why it works

In an availability cascade, the prevalence of worry is itself fed back into perceived probability: "everyone is concerned" becomes evidence that the concern is valid, which grows the concern further. This is social proof applied to risk — and it is exactly backwards as a probability signal, because social proof tracks salience, not base rate. Explicitly separating "widely worried about" from "empirically likely" breaks this feedback loop.

How to do it

1. When you find yourself thinking "everyone knows X is a major risk," flag this as a social-proof data point, not a probability data point.
2. Ask: "What is the actual incidence, independent of how worried people are?"
3. Treat popularity of worry as a variable that demands extra scrutiny, not as confirmation.

Evidence

Social proof effects on risk perception are documented in the social amplification of risk framework (Kasperson et al., 1988). The feedback from perceived social concern to individual concern is part of the cascade mechanism. (observational)

Sources

• Kasperson et al. (1988), "The Social Amplification of Risk: A Conceptual Framework," Risk Analysis

Common mistake

Using the absence of widespread worry as evidence that a risk is small — cascades can be slow to start; silence is also not reliable as a probability signal.