What to trust in a "post-truth" world | Alex Edmans

Episode Summary

Executive Summary: Alex Edmonds argues that in a post-truth world, fact-checking alone is insufficient: stories, facts, and even data can mislead if they are not representative or tested against rival explanations. Using Belle Gibson’s false cancer recovery story, he shows how confirmation bias distorts judgment and urges listeners to seek dissent, trust rigorously vetted expertise, and pause before sharing.

Main Topics: Belle Gibson and confirmation bias (Priority: 5/5): The talk opens with Belle Gibson’s viral cancer-recovery story to show how people accept narratives that fit what they want to believe, without checking truth. From post-truth to post-data (Priority: 5/5): Edmonds argues the deeper problem is not only false stories but our preference for single anecdotes over representative data. Bayesian thinking and rival theories (Priority: 5/5): He explains that evidence must support one theory while ruling out alternatives; data that merely fits a theory is not enough. How to avoid bias in decision-making (Priority: 4/5): Three practical steps are offered: seek dissenting viewpoints, listen to experts, and pause before sharing information. The role and limits of expertise (Priority: 4/5): Experts matter because they use evidence and peer review, but their claims still need scrutiny for credentials, rigor, and publication quality. Responsible sharing and evidence standards (Priority: 4/5): He warns that sharing unchecked claims spreads harm and recommends asking whether a claim is true, representative, supported, and robustly vetted.

Key Arguments: A story can be false, but even a true story may be irrelevant if it is only one anecdote and not representative. Confirmation bias leads people to ask whether data is consistent with a theory rather than whether it actually supports it over rival explanations. Data is not evidence unless it rules out alternative theories; evidence requires comparison against competing hypotheses. People overvalue vivid outlier cases and ignore the ordinary cases that make up the real pattern. To reduce bias, people should actively seek opposing views and create environments where dissent is encouraged. Expertise should be trusted more than gut instinct, but only when the expert’s credentials, methods, and publication venue are credible. Before sharing information, people should verify truth, scale, source quality, and whether they would believe the opposite result from the same authors.

Data Points: Belle Gibson app downloads: 200,000 in the first month - The Whole Pantry app gained rapid popularity after her story went viral. Cancer prognosis in the story: 4 months to live - Belle Gibson claimed doctors gave her a terminal brain cancer diagnosis. Treatment duration: 2 months - She allegedly underwent chemo and radiotherapy before claiming recovery. Peer-review rejection rate: 95% - Edmonds says the most elite academic journals reject most submissions. Study coverage timing: before publication - A November 2016 executive pay study was covered by newspapers even though they had not seen it yet. Word of the year: 2016: post-truth - He references the Oxford Dictionary’s selection to frame the cultural context.

Pivotal Quotes: "The biggest problem is not that we live in a post-truth world, it's that we live in a post-data world." — Alex Edmonds: He argues that people prefer isolated stories over representative evidence. "Data is just a collection of facts. Evidence is data that supports one theory and rules out others." — Alex Edmonds: He defines the difference between raw data and meaningful evidence. "Only if it's true can it be fact. Only if it's representative can it be data. Only if it's supportive can it be evidence." — Alex Edmonds: He closes with a hierarchy for evaluating claims before sharing or acting on them.

Implications: Listeners should treat anecdotes, headlines, and even studies skeptically unless they are representative, vetted, and compared against alternatives. Better decisions in health, business, and politics require dissent, expertise, and disciplined evidence standards.