EconTalk

In Defense of Intuition (with Gerd Gigerenzer)

December 29, 2025

Key Takeaways Copied to clipboard!

  • Intuition is defined as a fast feeling based on years of experience, which is not arbitrary, not a sixth sense, and available to everyone with domain expertise, contrary to historical gendered associations. 
  • The dichotomy between System 1 (fast/intuitive) and System 2 (slow/deliberate) thinking, often used to frame debates like those between Gigerenzer's and Kahneman/Tversky's research, incorrectly posits intuition and conscious thinking as opposing forces, whereas they often work together (e.g., intuition inspiring diagnostics). 
  • The 'bias bias' is the overemphasis on finding biases in human decision-making, where many claimed biases (like the hot hand fallacy or overconfidence) are actually ecologically rational heuristics that are beneficial in certain environments, or the result of flawed statistical interpretation by researchers. 

Segments

Defining Intuition and History
Copied to clipboard!
(00:01:04)
  • Key Takeaway: Intuition is a fast, unexplainable feeling derived from extensive experience, not an arbitrary decision or a sixth sense.
  • Summary: Intuition is defined as a feeling based on years of experience that rapidly enters consciousness without immediate explanation. Historically, intuition was wrongly gendered as female expertise versus male rationality, creating a dichotomy that persisted into the 20th century. This historical split is mirrored in the modern System 1/System 2 framework, which Gigerenzer argues misinterprets the relationship between fast and deliberate thinking.
Intuition’s Comeback and AI Limits
Copied to clipboard!
(00:04:32)
  • Key Takeaway: There is a growing respect for intuition, which is essential for innovation, and current Artificial Intelligence lacks the intuitive capacity to solve complex, uncertain human problems.
  • Summary: The perception of intuition is shifting positively, recognizing its necessity for innovation. Current AI, based on deep neural networks, excels in well-defined worlds like chess but struggles with uncertainty, unlike human intuition which relies on unconscious pattern recognition from experience. Claims that AI will solve complex social problems like poverty are likened to religious faith, ignoring the complexity demonstrated by projects like the Human Genome Project.
Critique of the Bias Bias
Copied to clipboard!
(00:19:09)
  • Key Takeaway: The ‘bias bias’ is the tendency to over-attribute biases to decision-makers, often overlooking that many commonly cited ‘biases’ are ecologically rational heuristics or stem from researchers’ statistical errors.
  • Summary: The bias bias involves seeing biases everywhere, often used to justify paternalistic policies like nudging. Many alleged biases, such as the hot hand fallacy, have been shown to result from researchers misunderstanding sample properties rather than errors by the decision-makers. Concepts like overconfidence or base rate neglect are not universally wrong but are ecologically rational, depending on the situation’s context.
Boosting Over Nudging
Copied to clipboard!
(00:28:24)
  • Key Takeaway: Boosting empowers individuals with risk literacy and understanding, contrasting with nudging, which is viewed as an anti-democratic form of paternalism that bypasses education.
  • Summary: Nudging, based on the premise that people cannot learn or deal with risk, justifies paternalism by subtly guiding choices. Boosting, conversely, aims to make people strong by teaching them concepts like Bayesian thinking or how to interpret probability statements (e.g., 30% chance of rain). True democratic participation requires people to understand information, making education superior to manipulation via defaults or framing.
Framing, Defaults, and Screening Deception
Copied to clipboard!
(00:35:32)
  • Key Takeaway: Framing effects, like using relative risk in medical communication, and default settings, like organ donation opt-out policies, often mislead the public because they ignore the underlying systemic or contextual realities.
  • Summary: The organ donation example shows that changing the default (opt-out vs. opt-in) does not increase actual donations if the complex organizational system required for transplantation remains unchanged. Similarly, communicating mammography screening benefits as a 20% relative risk reduction deceives women, as the absolute benefit is zero days of life extension on average, while ignoring the harms from false positives and subsequent testing.
Intuition in Morality and Science
Copied to clipboard!
(00:53:30)
  • Key Takeaway: Moral behavior is largely driven by intuitive, culturally absorbed rules that are later defended by conscious reasoning, serving the evolutionary function of bonding social groups.
  • Summary: Children absorb moral rules intuitively based on their upbringing, often defending these feelings later with explicit reasoning. Morality and religion evolved to bond social groups for survival against external threats. Scientists should defend core values by appealing to this shared social fabric rather than remaining timid or overly focused on individual logic.