You Are Not So Smart 073 - Bayes' Theorem
9 snips
Apr 8, 2016 
Neil A. Manson, a philosophy professor at the University of Mississippi, and Julia Galef, co-founder of the Center for Applied Rationality, delve into the nuances of belief and probability. They explore how beliefs are not simply true or false, but exist on a spectrum of certainty. The discussion centers around Bayes' theorem, illustrating its importance in updating beliefs with new evidence. They emphasize the need to embrace a 'grayscale' approach to thinking, where confidence in beliefs can shift based on context and information, helping listeners navigate uncertainty.
AI Snips
Chapters
Books
Transcript
Episode notes
Ignoring Base Rates Example
- People often ignore statistical base rates and rely on stereotypes when judging probabilities.
- This leads to worse decisions than if they used just the statistical facts alone.
Evolution of Fast Thinking
- Humans default to fast thinking because evolution favored quick decisions under pressure.
- Slow thinking, although better for accuracy, is avoided due to cognitive effort and time constraints.
Wartime Bayesian Thought Experiment
- Imagine spotting a distant enemy soldier knowing sharpshooters are 1% but far more accurate.
- Getting shot updates your belief that the soldier is likely a sharpshooter, illustrating Bayesian updating.
