Ep73 “The Dangers of Group Think on Decision Making” with Adi Sunderam

Quick Overview

Professor Adi Sundaram argues that group think is prevalent because people naturally seek confirmation for their existing beliefs, leading to the underestimation of true uncertainty, as demonstrated by the failure of many financial models and societal narratives to predict events like the COVID-19 impact.

Key Points: Group think is driven by the natural human tendency to seek confirmation for prior beliefs rather than rigorously testing them against all available data. The failure of many financial models to predict outcomes like the COVID-19 impact showcases the danger of group think, where models are often overfit to known data. Adi Sundaram highlights that people often favor explanations that confirm existing narratives (e.g., in political discourse or stock market analysis) over those that challenge them, leading to cognitive dissonance. The speaker suggests that a better approach involves forcing oneself to consider alternative explanations, even those with low probability, before settling on a conclusion. Social media amplifies group think by creating echo chambers where similar viewpoints reinforce each other, making people less likely to entertain contradictory evidence. Sundaram notes that in his research, he intentionally seeks out evidence that contradicts his initial hypothesis to avoid the trap of confirmation bias. The discussion references the high computational cost of rigorous model testing versus the ease of relying on simple, intuitive heuristics that confirm existing beliefs.

Context: This interview segment from the "All Else Equal Podcast," hosted by Jonathan Berrigan, features Adi Sundaram, the Willard Prescott Smith Professor of Corporate Finance at Harvard Business School. The discussion centers on the psychological phenomenon of 'group think' and its dangerous implications for decision-making, particularly in finance and broader societal contexts, emphasizing how the desire for cognitive comfort overrides rigorous, data-driven analysis.