Scientific consensus should guide decisions, but scrutiny is essential, especially in high-stakes scenarios. Not all scientists are objective; personal stakes, such as reputation, can impact their integrity and willingness to embrace dissenting views. A Bayesian approach recognizes that while the consensus is usually more reliable, individuals are encouraged to critically assess motivations behind scientific conclusions and their implications.
When physician Walter Freeman died in 1972, he still believed that lobotomies were the best treatment for mental illness. A pioneer in the method, he was a deeply confident and charismatic man who eagerly spread the technique in America, long after the rise of alternative treatments that were less destructive. Listen as journalist Megan McArdle and EconTalk's Russ Roberts discuss what McArdle calls the "Oedipus Trap": mistakes that no one can live with, even if they were innocently made, and how admitting such mistakes to ourselves is nearly impossible. They also discuss the complexity of the credo, "follow the science."