人性賽局: 哈佛大學最重要的行為經濟學課，驚人「隱藏賽局」完美解釋非理性行為

by Erez Yoeli, 摩西.霍夫曼, 趙盛慈 (translator)

★【哈佛經濟系×MIT史隆商學院】重磅經濟學家連手出擊★
★TED Talks影片超過240萬次觀看★
★《華爾街日報》熱烈推薦★
★ 台大國發所兼任教授、《賽局理論》作者周治邦 審訂、推薦★


◆為什麼幼兒園規定要罰錢，家長晚接小孩狀況更多？
◆為什麼獎金不一定能讓員工更努力工作？
◆奢侈品的價值，不在於它本身，而是能夠讓我們展現地位？

關於人性，傳統經濟學假設是理性的，行為經濟學則著眼於非理性的那一面，但兩端之間其實存在一個重要的知識領域：「隱藏賽局」。

不論是人的喜好、興趣，乃至「說不上為什麼」的直覺，還是炫富、利他、守規矩等社會現象，背後都有賽局邏輯，卻往往隱而不顯；
必須從主要獎勵、終極解釋、客位解釋3大角度切入，才能找出日常生活中的各種「隱藏賽局」。
作者舉出12種常見賽局，讓你看懂人類各種看似矛盾的非理性行為，其中無處不暗藏著精微的賽局計算。

/ 人性中的鷹鴿賽局：潛規則與制度的合理化 /
當人們習慣「誰當老鷹、誰當鴿子」，就會開始認同財產權、言論自由，甚至是厭女情結和加害者。
透過「主要獎勵」角度，看出隱藏的鷹鴿賽局，看懂人性習慣預設立場的利與弊。

/ 人性中的昂貴訊號賽局：醉翁之意不在酒 /
同樣

Genres: Economics, Psychology, Nonfiction, Science, Business, Sociology, Academic

332 pages, Kindle Edition

Published September 28, 2023

Reviews

[Maher Razouk · Rating 5 out of 5]

كيف طوّر الهنود ذوقاً للمأكولات الحارة جدًا؟ لماذا لم يفعل يهود أوروبا الشرقية الشيء نفسه؟ في عام 1998 ، اقترحت جينيفر بيلينج وبول شيرمان أن الإجابة تكمن في قدرة التوابل على تثبيط وقتل البكتيريا التي تسبب تلف الطعام. لقد اكتشفوا أن هذه القدرة كانت مفيدة للغاية في المناخات الحارة ، لذلك هذا هو المكان الذي نتوقع أن نجد فيه أشخاصًا تعلموا حب التوابل الحارة.
.
Erez Yoeli
Hidden Games
Translated By #Maher_Razouk

[Chris Boutté · Rating 5 out of 5]

A friend had a review copy of this book and kept tweeting out sections of it, and I had to have it. So, I reached out to get my review copy, and I binged it within a couple of days. I’ve been really interested in game theory for a while when it comes to human behavior and optimal ways to deal with people and make decisions. Unfortunately, game theory books can get extremely complicated and aren’t often written for laypeople. With this book, Yoeli and Hoffman did a great job writing a book that was accessible to people like me, but I also think it’s a great read for people more familiar with the topic than myself.

Game theory is often discussed when we’re talking about wars between nations or some other type of really serious situation. I’m more interested in our day-to-day relationships. This book did such an excellent job discussing how we evolved to cooperate but also to hold people accountable. It also dives into how we signal our status and position, and the authors give excellent theories and explanations for why we signal or do certain behaviors that are costly or seemingly have no benefit.

There were a few parts that went way over my head when they were explaining different research or game theory experiments, but it was mainly just the way the situations were set up. After it got passed the “If X does Y and Q is the situation” type stuff, they did an awesome job breaking it down and explaining the findings and potential logic behind the behaviors.

All-in-all, this is one of the best books I’ve read on the topic, and it’s even one of the better behavioral psych books I’ve read that explains seemingly irrational behavior. I really enjoyed it and will most likely be reading it again.

[Ozan · Rating 5 out of 5]

Very good book, comes as a saviour of Game Theory in the age of behavioural economics which shattered the strong rationality assumption that underlies GT. Behind the empirical anomalies, the authors show how relevant GT still is. A good read indeed.

[Graydon · Rating 5 out of 5]

Animal learning. We learn by doing and being rewarded. We also learn from others by asking and observing. We often learn in communities from those who we view as the most reliable on the subject.

Discussed spicy food preference and climate related health benefit. Point being that cultural preferences and practices are learned that correlate to health benefits, and these practices are followed despite the observers not knowing the benefits of these practices.

A lag is when a learned behavior has outgrown its benefit but is still observed. This is, sexual practices despite birth control usage, spicy food despite large scale refrigeration, and status seeking despite increase in status not leading to more mates due to single mate commitment.

A spillover is when the learned behavior spills over. This is a dog humping the owners leg, or a sense of fairness which evolved to protect us from being taken advantage of or walked all over- which in modern day life might lead us to behave irrationally, or against our own self-interest.

Emic vs etic explanations
Ultimate vs proximate explanations
Fisher sex ratio explanations
Nash equilibrium

Separating Nash equilibriums
Costly for the signaler, less costly to send for some (the more fit party).
Birds with long tail feathers, cost of the long feathers vs the signal of fitness.

Senders bury costly signals. Anonymous donors. They can afford to pay the cost of some recognition, and benefit an increased reward when the withheld information is discovered to select parties.

Biased revelations, selectively sharing information spin
Biased research of evidence, spin intensive search of supportive evidence.
Confirmatory testing

Evidence games
Persuasion
Motivated reasoning

Overconfidence
Asymmetric updating
Asymmetric research

Attitude polarization

Internalized polarization
Diagnostic and on diagnostic evaluations and evidence

Motivated beliefs and internalized persuasion

Internalized beliefs reflect motives. BP oil executives. Defense and persecution lawyers.

Repeated prisoners dilemma
Nash equilibrium
Sub game perfect
Reciprocal altruism, conditional cooperation

Trench ware fare, repeated battles, high delta, opposing armies cooperated and pretended to attack each other
High Deta means many repeated interactions
Diatic relationships

Sub game perfect norm enforcement
third-party and hire order enforcement
Paid to punish those who have done some thing to violate the public good despite it not impacting you directly

How can we make marriage and ask equilibrium? If the Nash equilibrium is to stay faithful, and use reciprocity or conditional cooperation to keep the marriage going this might be the way.

People are motivated to third-party punish by the threat of higher order punishment. This was proven that they would third-party punish even though they were not convinced that the person had done anything wrong, if the threat of higher order punishment was there.

Altruism:
Observability
Expectations of behavior

Use these strategies to make people comply to common good. Like the pirate code.
Norms for good relations and cooperation
Outlaw- protection of law removed
Incentivize norm following

Categorical versus continuous norms. A categorical 9 pounds of food by chemical warfare and does not measure against continuous variables like number of deaths or suffering.

State and signal structure
States high and low states, peacock fitness, fit or unfit.
Priors Indicate the odds that his state will happen before players know.
Players get signals about the state, are you signals are noisy and might have error

State signal structure, what is the combination of signals about the states.
Risk dominant indicates the safer bet in the coordination game.
Nature determines the state then players
Sense the signal. Then players play the coordination game.
Uncorrelated asymmetry. These are things that don’t affect the value or the likelihood of an outcome. But they do influence the way players will play the game.

Continuous versus discreet signals from nature. Which are you able to enforce norms about.

Trolley problems. Omission versus commission. It’s viewed as less bad if omission.

Human relationships, four bins..Communal sharing, in this burn your focus on similarities of you is the same and ignore differences.
Authority ranking relationships.Hierarchy. Those above have prestige those below are protected.
Equality matching relationships.
Tit for tat, they are concerned if the relationship is balanced.Market pricing relationships.


To punish based on omission or inaction intent is needed- and this is difficult to coordinate around
When coordinating is not required intent can be considered more

Avoid the ask, Strategic ignorance

Use of innuendo and indirect speech, like shall we go upstairs for a drink, enables us to get our first order needs communicated well obscuring higher order visibility of our request. This enables plausible deniability while still requesting or acting in the potentially antisocial way. Or sanctionable way.

Apologies are important because they change expectations to the players about what equilibrium they are in and can expect. In a sub game perfect cooperation game players must exploit failure to punish Transgressions.
. Also this is why you must punish or else you are incentivizing exploitation.
When the transgressor benefits from there transgression the apology must be costly, or else you can expect the transgressions to continue.

Costly signaling. Using a waste for signal, Provides useful information if I can differentiate you from my less fit individual. The signal or will be rewarded.

Functions of passions. Passions function to signal and incentivize us to invest in skills and abilities,That are likely to reap social and material benefits later on.
You won’t develop a passion for something that won’t be socially rewarding.
The amount of passion you develop also depends on how weak you are in other areas.
Passion is more likely develop in an arena where there are superstars. When the rewards go to a select small group all or nothing, passion is much more helpful. Art sports law.

Grit.

If you get paid for a task you loose motivation. Needing to lay is a Signal of lack of intrinsic value. Crowding out intrinsic value.

Flow
Induce flow in the workplace, and become more productive.
Clear goals immediate feedback
And balance between opportunity and capacity.


Grits passion and Flo are likely when there’s an opportunity and the probability of success. Disconnects with learned helplessness if you do not have the probability of success in your feedback that indicates this you will be less likely to try..

Game theory and the hidden accounting system of primary rewards.

[Jukka Aakula · Rating 3 out of 5]

The beginning and the idea of hidden games and especially chapter 5 on shared expectations and hawk dove were very interesting and important. Certainly very much worth reading. Chapters 4 and 6-12 were partly very familiar to me but also quite good material. But in chapters 13 - 14 I lost interest - too many stories and too many details to me.

[Roland Andrews · Rating 5 out of 5]

Very valuable.

[Ben Sicnolf · Rating 4 out of 5]

I could talk game theory stuff for hours so I enjoyed this - reason for 4 stars - but I feel like this book was not as good as other stuff I have read involving similar things. The real life examples were good illustrations and fun but some of the technical jargon lost me.

[Richard · Rating 3 out of 5]

Nice introduction to Game Theory. Loses a couple of stars as it gets very technical and academic in sections and as a result becomes a challenging bedtime read.

[Alice Wardle · Rating 3 out of 5]

This book presents itself as being in the popular science genre - suitable for a general audience of people interested in knowing more about Game Theory. As someone who has read several books about Game Theory and its relevance to psychology, I thought I'd read this book to try to understand complex human behaviours through the application of Game Theory.

The authors', Erez Yoeli and Moshe Hoffman, tone throughout this book was casual and friendly, perhaps too much so. They threw in some jokes randomly throughout the book, some of which made me laugh, while others sounded forced and like the writers tried too hard to be funny and entertaining. There is also repeated use of the colloquialism 'ain't', which didn't seem to fit or be appropriate for the text; some components felt out of place, including the algebraic formulations and Bayesian Statistics at the end of each chapter. The book went from a well-detailed and entertaining explanation of social behaviour, and all of a sudden, Greek letters and more-than (>) and less-than (<) signs were propelled towards my eyes like darts. I wasn't told I needed a introductory course in statistics before reading this book! I was told "ρ as ρ = (r (1 - ε) - (1 - ε)2) / (1 - ε - (1 - ε)2)." This isn't...whoops, I mean 'ain't'...good for someone who didn't do an advanced Maths course.

The book uses biological, evolutionary, and mathematical perspectives to understand social behavioural anomalies. The book has some very genuinely valuable and interesting chunks of information, but I'd only suggest reading it if you have some awareness of Bayesian Statistics or mathematics, otherwise reading will likely be tedious. Honestly, if the mathematical sections would have been more accessible and understandable to math-novices, then this easily would have gotten a better review.

[Teo]

2023.09.07–2023.09.14

Contents

Yoeli E & Hoffman M (2022) (10:41) Hidden Games - The Surprising Power of Game Theory to Explain Irrational Human Behavior

01. Introduction

02. Learning
• Reinforcement Learning
• Social Learning
• Lags and Spillovers

03. Three Useful Distinctions
• Primary versus Secondary Rewards
• The Proximate-Ultimate Distinction
• The Emic-Etic Distinction

04. Sex Ratios: The Gold Standard of Game Theory
• Nash Equilibrium
• Arriving at a Nash Equilibrium without Conscious Optimization
• How to Tell if a Game Theory Explanation is the Right One
• Comparative Statics
• The Optimization Process Affects the Interpretation

05. Hawk-Dove and Rights
• The Fierce Speckled Woods of Oxfordshire Forest
• Property Rights in Humans
• Generalizing beyond Property Rights
• Apologies
• Should You Be More Assertive?
• Stockholm Syndrome and Internalized Racism and Sexism

06. Costly Signaling and Aesthetics
• The Game
• Some Evidence from Sunbirds
• Luxury Goods
• Other Signals of Wealth (and Occupation)
• Signaling Other Attributes

07. Buried Signals and Modesty
• Anonymous Giving
• Overeagerness
• Shibui
• Our Explanation

08. Evidence Games and Spin
• Biased Revelation
• Biased Search
• Confirmatory Testing
• The Game Theory
• Model 1: Revelation
• Model 2: Search
• Model 3: Testing

09. Motivated Reasoning
• Internalized Persuasion
• Some Evidence
• Alternative Explanations

10. The Repeated Prisoner’s Dilemma and Altruism
• The Repeated Prisoner’s Dilemma
• Equilibria of the Repeated Prisoner’s Dilemma
• Subgame Perfect Equilibria
• The Necessary Features of Cooperation
• Some Evidence

11. Norm Enforcement
• Observability
• Expectations
• Limits on What’s Enforceable
• Practical Advice

12. Categorical Norms
• State-Signal Structures
• Adding a Coordination Game
• Continuous versus Discrete Signals
• Unraveling
• Key Assumptions
• Back to Our Puzzles
• Evidence
• Unraveling in Practice
• Bonus Application: (In)Efficient Giving

13. Higher-Order Beliefs
• The Core Idea
• Observability
• Shared Signals
• Plausible Deniability
• Higher-Order Uncertainty
• Mapping to Our Puzzles
• Symbolic Gestures
• Omission-Commission Distinction
• Indirect Communication

14. Subgame Perfection and Justice
• Duels
• Moral Luck
• Full-Blown Responses to Minuscule Transgressions

15. The Hidden Role of Primary Rewards

Acknowledgments
Notes

[Minervas · Rating 4 out of 5]

Economists often use game theory to explain patterns of behavior, but the authors of this book take the approach one step further. They use this mathematical framework to explore cultural norms, moral instincts, and even deeper inner experiences of human beings.

Throughout the book, the authors present ideas that are both interesting and convincing. After visiting their website, I found that several of their arguments are strikingly original. For example, they propose that motivated reasoning arises from an internal need to persuade others, rather than merely from the pursuit of comfort (“hedonic hacking”). This argument is supported through thoughtful comparisons and experimental evidence. I also appreciated the authors’ explanation of why social norms tend to be categorical. In short, categorical signals make coordinated punishment easier, which in turn sustains norm enforcement. This highlights how information structures underpin behavior and even evolved instincts.

Some points, while intriguing, felt incomplete. For instance, the authors suggest that modesty serves as a subtle way to signal one’s merits. While plausible, I think modesty might also function as a way to rein in runaway signaling competitions. I’d liked to see more discussion of such alternative interpretations.

lastly, I found parts of the last chapter far less compelling. The “broad-stroke” explanations of passion, grit, flow, and sense of meaning struck me as shallow and unconvincing. This chapter is the main reason I’m withholding a five-star rating.

Overall, this is a typical book written by economists—sometimes brilliant and profound, sometimes surprisingly narrow and naive. Still, if you read it with discernment, setting aside the weaker ideas, it’s well worth your time.

(Edited with AI assistance)

[Jason · Rating 2 out of 5]

Dear author:

The good: you've introduced us to the basics of game theory. Good for us.

The very bad: you've completely misapplied the ETIC nature of this logical model of decision-making, in an attempt to explain psychological and learning processes. Processes which *require* EMIC ways of knowing and being in the world.

Example: One canNOT be taken seriously when attempting to justify game theory (which you do ceaselessly and withOUT data, by the way), to explain Csikszentmihalyi's work on Flow. One canNOT, particularly when invoking experiments about learned helplessness in dogs (a very ugly, negative reinforcement exercise). FFS, the dogs didn't need the motivation found through a Flow state to somehow ameliorate their learned helplessness.

That aside, there are leaps of logic EVERYWHERE. Game Theory canNOT predict outcomes and while it does well at explaining large-scale decision-making in warfare and economics, it does so because one doesn't always look at the particulars of a group--merely the over-riding ETIC decision-making done by large groups to maximize chances of WINNING.

PLEASE READ CRITIQUES OF GAME THEORY SCHOLARSHIP before attempting to misapply this body of work.

[April · Rating 3 out of 5]

📚 Book #79: “Hidden Games” by Erez Yoeli and Moshe Hoffman

📕 This book is an explanation of game theory written by two MIT economists. Yoeli and Hoffman talk through several examples of behaviors that seem irrational at first, but can be understood through further investigation of the patterns. These examples are challenged through sets of questions that help determine whether a behavior is a Nash equilibrium for that experience -- making it a logical place for the behavior to have settled.

⭐️⭐️⭐️ 3/5: I didn't end up enjoying this read as much as I had expected. Game theory is a tricky subject to write on. Much of this book was too technical and math-y for the average person to follow. Much of it was also boring, repetitive, or common sense to people that know anything about game theory or behavioral psychology already. Perhaps game theory just wasn't as interesting of a topic to me.

🤓 You should read this if you're curious about game theory! This is a good intro if you don't know about game theory, but are willing to sit through quite a few equations and science-y explanations.

[Jason Schlosberg · Rating 3 out of 5]

The narrative explaining each game's concepts is exceptional and clear. However, this text suffers greatly from developing from the beginning a clear and organized roadmap giving an overall sense of game theory's meaning and parameters. As a result, it appears the authors jump arbitrarily from concept to concept without sufficiently tying each one into the main premise. Moreover, while it is understandable that the author's would want to provide some mathematical functions inherent to game theory, I did not find it helpful, since it lacked sufficient explanation and remained too obtuse for this lay reader.

[Robin · Rating 2 out of 5]

Never let an or a pair of economists convince you of anything, they will try to convince you of things that don't exist & well obfuscate the understanding of those that do. In fact almost all economists ( not all but most including those penning editorials & making guest appearances on TV) are better understood as fiction peddlers , the crucial difference between them & other fiction peddlers is that neither they themselves nor the audience at large, including policy makers, seem to be aware of this fictional nature of the concepts, stories & understanding peddled.
This is equally true even if the economists happen to be from MIT.

[Niek · Rating 2 out of 5]

There is something off about this book that frustrated me. At times, it is like the authors underestimate their readers' intelligence, e.g., by building the countless examples they use to drive home the point they ask interesting questions. Then, when answering their questions all of the sudden they assume their audience to be math or psychology students.

I am not even going to start about the overlap with other with other pop psychology books (I don't get what learned helplessness, grit, and flow are doing in this book). There also is a lot of referring to past and future chapters I did not care for (e.g., "we are going to come back on this later", "as well explained in chapter 5").

[Homa]

Hidden Games tries to show that behind many of our seemingly irrational behaviors lies a hidden logic rooted in game theory. It’s an intriguing idea that we play for survival, trust, or social status, even when we think we’re simply acting on emotion.
The book’s strength is its fresh perspective on human behavior, and it makes you look at yourself and others with sharper attention. Still, I think it sometimes explains everything through the lens of strategy and self-interest, leaving little room for the emotional and cultural complexity that shapes people.
Overall, it’s a smart and thought-provoking book, but best read with a skeptical mind, not as the final truth about human nature.

[Florian, Daniel · Rating 4 out of 5]

Why do some conflicts that seem trivial at first glance escalate? Why is history so decisive for a state's territorial claims? Why is "spin" so important in politics and how can "spin" be debunked? Game theory has enormous potential for analysing political processes, but often remains abstract and seems to ignore "irrational" aspects of politics such as ideology and emotions. In their short book, Hoffmann and Yoeli address precisely the question of how bounded rationality can be incorporated into game-theoretical models, thus closing an important gap between theory and practice.

[notirene · Rating 2 out of 5]

This book took me a while to get through not only because of the theoretical content but also because every mathematical model described involves probability, logic, and way too many variables. All the storytelling with real world examples was enjoyable and everything else took me too long to digest. I wouldn’t recommend this read to anyone else unless they were a mathematician, statistician, or heavily enjoy reading about people’s behaviors explained away with convoluted verbiage and punnet square looking diagrams. Lol

[Andrew Jiang · Rating 3 out of 5]

Nothing too revolutionary in this book — it contains the same tried formula of pop psychology books that continue to proliferate in the world. I do appreciate more of the mathematical models that are built out as that usually is lacking from these kinds of books, but otherwise most of the content points out seemingly intuitive results.

There’s also some concerns with tone, especially when the book discusses quite contemporary topics.

[Kat · Rating 3 out of 5]

3.5 Stars
Game theory is super interesting and this book explains it quite well in my opinion. The basics are easy enough to understand it. The authors put the theory into perspective by giving many examples on human behavior. Some chapters went into more detail and since I was reading the audiobook it was hard to follow equations and parts therefore went over my head. Overall a solid read.

[Megan · Rating 2 out of 5]

Overly simplified, but not in a useful way. Much of the concepts were presented for a beginner, but used advanced logic to explain the concepts. This book was either too patronizing for a non-beginner, or too complicated for a beginner. The focus was much more on social evolution/anthropology than game theory.

[Nick Lucarelli · Rating 3 out of 5]

A fair book to add to the growing behavioural psychology compendium (/bandwagon?). Doesn't add a heap aside from some great insights into why signals of wealth/power etc have evolved and why they still persist, as well as the usual anecdotes and marshmallow test pop psych history. Even attempts to distill it down into maths which was far beyond me.

[Tom · Rating 1 out of 5]

It was an interesting introduction to game theory, but was very poorly written, and somehow both over-simplified and over-complicated. A very odd read and wouldn’t recommend it over any other game theory book.

[Alex]

https://www.linkedin.com/posts/erez-y...

[Sanchita Das · Rating 5 out of 5]

Very clear and well written. Underlines cold logic in determining the outcomes of social life - Spots simple game theoretic models at play in many daily interactions of rational/ irrational agents.

[Damien A. · Rating 4 out of 5]

Fascinating. I've always loved game theory. However, the math is quite complicated

[Ayesha Mian · Rating 4 out of 5]

Good read on game theory and how that can explain human behavior. Inspired me to read more books on game theory.

[Michael Zhou · Rating 5 out of 5]

Interesting dive into how game theory can explain the oddities of human behavior-- some of the chapters can be a bit technical but very insightful nonetheless!

[C · Rating 2 out of 5]

Gems are few and far between. Deep into game theory down to including math formulas. Largely dry.