
Show Notes
Welcome to a five-episode special series built around the first Annual Review of Psychology chapter on wisdom since 2011.
Episode 3: Measuring Wisdom (with Judith Glück, Monika Ardelt, and Eranda Jayawickreme)
Can we actually measure wisdom, or are we just measuring how wise people think they are? Judith Glück, Monika Ardelt, and Eranda Jayawickreme join Igor and Charles to explore what happens when we try to turn wisdom into something we can quantify. Judith reveals the surprising differences between questionnaires, performance tests, and assessments of people in real situations, Monika makes the case that wisdom involves a stable quality of character built around reflection, cognition and compassion, and Eranda asks whether we might be better off thinking of traits as patterns of behaviour that change from one situation to another. Along the way, Igor and Charles confront a fundamental question about what a wisdom score is actually telling us, and discover that the answer may be less about choosing between traits and states than understanding how they fit together. But if wisdom is a pattern of states, can we change that pattern? Special Guests: Eranda Jayawickreme, Judith Glück, and Monika Ardelt. Links: Navigating the Unknown: The Sociocultural and Metacognitive Architecture of Wisdom (accepted) | Annual Review of Psychology, Volume 78 - Igor Grossmann and Nic M. Weststrate
More from On Wisdom
Welcome to a five-episode special series built around the first Annual Review of Psychology chapter on wisdom since 2011.
Welcome to a five-episode special series built around the first Annual Review of Psychology chapter on wisdom since 2011.
Welcome to a five-episode special series built around the first Annual Review of Psychology chapter on wisdom since 2011.
What can insights from the psychology of technology teach us about wisdom in the age of AI? In this special follow-up episode, Igor and Charles are joined by Steve Rathje to explore how classic ideas like the Turing Test hold up now that AI can talk compellingly about human wisdom. Steve unpacks what today’s generative models are actually capable of, Igor is intrigued by how quickly the line between human and machine reasoning seems to be blurring, and Charles realises that telling human insight from machine insight isn’t nearly as straightforward as he'd hoped. The trio also reveal the results of our listener poll — who sounded the wisest, and was the audience able to spot the AI? Welcome to Episode 67. Special Guest: Steve Rathje. Links: Steve Rathje's Site: