Acronym

New York Times science writer John Tierney wrote a column and a couple of interesting blog posts (here and here) on a sociological phenomenon known as a "cascade." In the column, he uses the evil of the fatty diet as the example, saying a few highly regarded scientists first made the argument that fatty diets led to shorter lives. Turns out just about all rigorously researched studies since the initial claims could find no correlation, yet the myth that a connection between an early demise and a fatty diet is scientifically accepted persists today.

The reason, Tierney says, is a cascade. (Disclosure in case you don't go to my source material: Tierney's column is about a book, Good Calories, Bad Calories by Gary Taubes.)

The connection I want to make is how Tierney dumbed down the explanation of a cascade. He used the show Who Wants to Be a Millionaire?, noting that when asked, the audience usually arrived at the correct answer. The reason is because they are all voting essentially in isolation. If instead each audience member was asked one after another, the answer would likely be vastly different as those who were unsure or mostly unsure would choose who everyone else was choosing. Everyone will recognize this a classic groupthink behavior.

Groupthink itself generally leads to poorer decisions than a situation in which there is no groupthink. Where this gets even more dangerous is when the groupthink leads to, for example, an entire association putting its reputation behind faulty conclusions. Now we have the beginnings of a reputational cascade. A scientist in high regard publishes a finding that is then endorsed by an organization, which leads to more endorsements and so on. In Tierney's example, it's the Surgeon General who promulgates the myth, making it seem almost unimpeachable. Politicians and organizations don't like going back on their decisions, but I think organizations have to be more open to skepticism without summarily dismissing it. It's ok to be wrong, as long your intentions are good and you continually try to be as correct as you can.