Simple Heuristics That Make Us Smart

Citation

Gigerenzer, G., Todd, P. M., & the ABC Research Group. (2000). Simple Heuristics That Make Us Smart. Oxford University Press.

Chapter Summary

The book has sixteen chapters in six parts and about 416 pages. The authors are members of the ABC Research Group (the Center for Adaptive Behavior and Cognition at the Max Planck Institute for Human Development), and each chapter is written by a subset of the group. The central claim, as the authors' own précis in Behavioral and Brain Sciences puts it, is that "fast and frugal" heuristics (simple rules in the mind's adaptive toolbox) let living organisms and artificial systems choose quickly with little information by exploiting how information is structured in particular environments. Heuristics are built from simple blocks that control information search, stop search and make a decision, and they can perform comparably to more complex algorithms, particularly when generalizing to new data.

Part I: The Research Agenda (Chapter 1)

Chapter 1: Fast and Frugal Heuristics: The Adaptive Toolbox (Gigerenzer and Todd) sets out the programme: a notion of rationality that fits real minds with limited time and knowledge, instead of the unbounded rationality assumed in much of cognitive science and economics. It is the framing chapter for everything else.

Part II: Ignorance-Based Decision Making (Chapters 2–3)

Chapter 2: The Recognition Heuristic: How Ignorance Makes Us Smart (Goldstein and Gigerenzer) presents the idea that recognizing one option and not another can be enough to decide. Chapter 3: Can Ignorance Beat the Stock Market? (Borges, Goldstein, Ortmann and Gigerenzer) tests the idea on stock picking.

Part III: One-Reason Decision Making (Chapters 4–8)

The longest part develops heuristics that decide on a single good reason. Chapter 4: Betting on One Good Reason: The Take The Best Heuristic introduces the best-known of them. Chapter 5: How Good Are Simple Heuristics? and Chapter 6: Why Does One-Reason Decision Making Work? A Case Study in Ecological Rationality test accuracy against more complex methods and ask under what structure of the environment simplicity succeeds. Chapter 7: When Do People Use Simple Heuristics, and How Can We Tell? moves from performance to the psychology of when people use them, and Chapter 8: Bayesian Benchmarks for Fast and Frugal Heuristics compares them with a Bayesian network.

Part IV: Beyond Choice: Memory, Estimation, and Categorization (Chapters 9–11)

Chapter 9: Hindsight Bias: A Price Worth Paying for Fast and Frugal Memory treats a familiar bias as a by-product of an efficient memory process. Chapter 10: Quick Estimation: Letting the Environment Do the Work and Chapter 11: Categorization by Elimination: Using Few Cues to Choose extend the approach from choosing between options to estimating quantities and sorting objects into categories.

Part V: Social Intelligence (Chapters 12–14)

Chapter 12: How Motion Reveals Intention: Categorizing Social Interactions, Chapter 13: From Pride and Prejudice to Persuasion: Satisficing in Mate Search and Chapter 14: Parental Investment by Simple Decision Rules apply heuristics to inferring intent from movement, to searching for a mate and to decisions about investing in offspring.

Part VI: A Look Around, A Look Back, A Look Ahead (Chapters 15–16)

Chapter 15: Demons Versus Heuristics in Artificial Intelligence, Behavioral Ecology, and Economics compares the heuristics approach with optimization-based ("demon") models in three fields. Chapter 16: What We Have Learned (So Far) (Todd and Gigerenzer) summarizes the programme.

Key Concepts

• Fast and frugal heuristics / adaptive toolbox: simple decision rules, in Chapter 1's terms, that a mind holds for use in different tasks.
• Recognition heuristic: deciding on the basis of recognizing one option and not another (Chapter 2).
• Take The Best: deciding on the single most valid cue that distinguishes the options (Chapter 4).
• Ecological rationality: the match between a heuristic and the structure of the environment (Chapter 6).
• Satisficing: accepting an option that meets an aspiration level, applied to mate search in Chapter 13.
• Categorization by elimination: classifying with a few cues (Chapter 11).
• Hindsight bias as a by-product: Chapter 9's reinterpretation of a classic error.

Critical Analysis

Strengths

• A unified programme. The same building blocks (search, stopping and decision rules) are traced through choice, estimation, categorization, memory and social decisions, so the book reads as a sustained research argument and not a collection of unrelated studies.
• Computational testing. The précis describes developing computational models of heuristics and testing them with simulations and experiments, and the chapters compare heuristics against methods such as multiple regression, a Bayesian network and a neural network.
• Range of domains. The Part V chapters on motion, mate search and parental investment take the idea beyond laboratory judgment tasks to topics from behavioral ecology.
• Clear programme statement. Chapter 1 and the closing chapter, "What We Have Learned (So Far)," frame the approach for readers new to it.

Limitations

• Length and repetition. J. Marty Anderies, reviewing the book in Conservation Ecology (2001), found it too long, with many examples sharing the same basic structure, in which heuristics do well on coarse-grained predictions.
• Open questions about scope. The same review argues that the book does not present examples in which heuristics perform very poorly, makes too much of the distinction between heuristics and constrained optimization, and does not establish that heuristics are better models of behavior than complex alternatives. These are one reviewer's views, but they identify where to read critically.
• Date and audience. It dates from the end of the 1990s, and the publisher's description addresses researchers in cognitive and evolutionary psychology, cognitive science, economics and artificial intelligence. It is a research monograph, not a first text on judgment and decision making.

Where It Fits in the Course

This is the research monograph in the judgment and decision-making list, and the main statement of the fast-and-frugal heuristics programme. Hastie and Dawes's Rational Choice in an Uncertain World and Evans's Thinking and Reasoning: A Very Short Introduction are introductions, Bazerman and Moore's Judgment in Managerial Decision Making is a short textbook on the heuristics-and-biases tradition, Baron's Thinking and Deciding is a comprehensive text, and Kahneman's Thinking, Fast and Slow and Thaler and Sunstein's Nudge are books for general readers.

Read Gigerenzer and colleagues after one of the textbooks, as the main counterpoint to the heuristics-and-biases view: where Bazerman and Moore treat heuristics as sources of systematic error, this book asks when they are accurate and why. Reading Chapter 1 beside Kahneman's account gives a quick sense of the debate.

How to Use This Book

The book is for graduate students and advanced undergraduates in cognitive and decision psychology, and for researchers in economics, behavioral ecology or artificial intelligence. A course in judgment and decision making or cognitive psychology is assumed, and basic familiarity with probability and statistical models helps.

A practical plan:

• Core: Chapters 1, 2, 4, 5 and 6 for the programme, the two main heuristics and the idea of ecological rationality; Chapter 16 as a summary.
• Then choose by interest. Chapters 7 and 8 for the psychology and statistical benchmarks; Chapters 9–11 for memory, estimation and categorization; Chapters 12–14 for social and biological applications.
• Read critically: Chapter 15 sets the approach against optimization models, and a published review such as Anderies's is a useful counter-reading.
• Work through one heuristic by hand on a small data set to see how cue order, stopping and the decision rule operate.