Game Theory Research Suggests We Can't Ever Fully Learn From Our Mistakes
We like to think we're superior rational creatures, but research reminds us that we're also selfish and emotional animals that don't always learn from our mistakes.
Hiya!
The Enlightenment, which occurred during the 17th and 18th centuries, led us to believe that humans are, by nature, rational creatures. And that our rational, logical mind separates and elevates us from the rest of the animal kingdom. There may be a seed of Truth buried in there somewhere, but just because we’re capable of rational thinking doesn’t mean we utilize it.
We can be compassionate and ambivalent toward others, but we can also be apathetic or even cruel when it serves us. Similarly, thinking about things with a strictly logical lens is shockingly tricky when emotions are involved.
Still, we tend to focus on the qualities we take pride in, like rational thinking, while keeping our more animal instincts in the shadows. Consequently, many social theories about Human behavior rely on these assumptions, but a growing pile of scientific research is holding up a mirror.
Game Theory
I didn’t know much about game theory before writing this article, but now that I’ve researched it, I understand why people think and treat managing a society (or business) like a game. According to Investopedia:
“Game theory is the study of how and why individuals and entities (called players) make decisions about their situations. It is a theoretical framework for conceiving social scenarios among competing players.”
And that, “In some respects, game theory is the science of strategy, or at least of the optimal decision-making of independent and competing actors in a strategic setting.”
Game theory is versatile and can be applied to various fields to test multiple situations and predict their most likely outcomes.
The theory goes back to 1944 when mathematician and physicist John von Neumann collaborated with economist Oskar Morgenstern to create and publish A Theory of Games and Economic Behavior, which argues that all economic situations can be defined as the outcome of a game between two or more players.
There are two primary elements involved in game theory: players and strategy. Players are considered rational decision-makers in a game, and strategies are the actions the players take depending on the circumstances that arise in the game. The strategies most players utilized were typically based on the player’s self-interest and what’s expected of the other players.
Other elements of a game
There are many variations of game theory, including simultaneous/sequential, cooperative/non-cooperative, and zero-sum/non-zero-sum. But regardless of the setup, they must each accord to a couple of standards.
To qualify as a game, all players must understand the rules, the strategies available to them to win, and the reward of winning. A game ends or is won when a player reaches equilibrium — when they make a decision and reach an outcome. A common assumption is that the players will behave rationally and make choices based on self-interest.
Stanford Encyclopedia of Philosophy says the “Prisoner’s Dilemma” is the most well-known example of game theory:
Imagine two people are arrested for a crime. The police have no hard evidence to convict them, so the officials separate the two suspects to interrogate them in different rooms, with no way for the suspects to communicate with each other. Then, the officials present four deals:
They will each receive a three-year prison sentence if they both confess.
If Suspect #1 confesses but Suspect #2 doesn’t, then Suspect #1 will go to jail for one year, and Suspect #2 will get five years.
If Suspect #2 confesses but Suspect #1 does not, Suspect #1 will get five years, and Suspect #2 will get one year.
If neither confesses, each will serve two years in prison.
According to game theory, the best strategy is for neither suspect to confess so they both receive a two-year sentence. But without the ability to communicate, neither suspect can know the other’s strategy, and without knowing for sure what their partner will decide, it’s more likely that each suspect will make the best decision for themselves but worse for them as a team.
The diversity of game theory is expansive and can be applied far beyond dilemma games. Examples can be used in everyday life, from Poker and Rock, Paper, Scissors games to conflict analysis, war and business strategies.
Economic Theories
I’ll be honest: economics was an elusive and vague concept I struggled to understand for a long time. But then I learned it’s really just a social science that studies how people respond to different situations and how they might respond to changes.
The American Economic Association says economics can be defined in “a few different ways,” but summarizes that,
“[Economics is] the study of scarcity, the study of how people use resources and respond to incentives, or the study of decision-making. It often involves topics like wealth and finance, but it’s not all about money. Economics is a broad discipline that helps us understand historical trends, interpret today’s headlines, and make predictions about the coming years.”
It also states that economics covers many situations, from small to large. Microeconomics focuses on individual decisions, such as families’ medical debt. Meanwhile, macroeconomics focuses on larger populations, like sovereign debt.
Most current economic game theories generally assume people make rational decisions based on lessons they’ve learned from previous experiences and that most people utilize this knowledge until they level out to a stable state of consistently maximizing their earnings.
But this assumption depends on the fact that humans are, on a foundational level, rational decision-makers. In other words, it assumes we, as a species, typically favor and evaluate known information, often disregarding our emotional reactions, to make our decisions.
However, research, especially regarding marketing and advertising, suggests the complete opposite — that our decisions are often emotional responses.
Physicist Jérôme Garnier-Brun at École Polytechnique in France was also surprised by the continued economic assumptions about human behavior because it conflicts with what physicists observe in nature.
New Research
In nature, most interactions — right down to the molecular level — typically end in chaos rather than stability. So, Garnier-Brun and his colleagues developed a mathematical game featuring hundreds of theoretical players representing a large economy to mathematically test whether economists are right in assuming that learning from the past helps people avoid chaos.
Each of the hundreds of theoretical players in the game could choose between two actions — such as buying or selling a stock — and interact with each other. Each player's decisions were influenced by their previous experiences, meaning they could learn from them.
The researchers could also adjust certain variables, including how much a player’s past experiences influenced their future decisions and how cooperative the players were with each other when interacting. For their research, Garnier-Brun and his colleagues set variables reflecting statistical physics to model different game scenarios.
If the economists are correct, then most scenarios would result in the virtual players eventually settling into a stable state and winning the game, with only some scenarios always resulting in chaos — meaning the players would be incapable of learning how to optimize their choices.
But that’s not what the results showed.
The Results
The researchers found only a single scenario in which all virtual players in the game performed “optimally” — when a “centralized, omniscient agent endowed with enormous computing power” was calling the shots. Otherwise, regular players only reached satisfactory states, meaning they satisfied the bare minimum expectations but not much more.
The players learned more than they would have at random, so learning from previous experiences had at least some influence, but they still learned less than they would have if they could optimize their performance.
In all, using a mixture of simulations and analytical arguments, the researchers concluded that long-term memory of past beneficial experiences can lead toward stabilization, but overreacting to a recent event typically ends in chaos.
They also found that non-reciprocity (increased competition) destabilizes fixed points, resulting in chaos. Paradoxically, some randomness in the learning process allowed the virtual players to make better collective decisions.
The team found one somewhat more positive outcome: The learning process “allows cooperative systems to coordinate around solutions with rather high (but markedly sub-optimal) average reward.”
One of the researchers, French physicist Jean-Philippe Bouchaud at École Polytechnique, told Karmela Padavic-Callaghan of New Scientist:
“The idea that people are always making complicated economic computations and learn how to become the most rational agents, our paper invites everyone to move on [from that].”
He says the game model his team invented is too simple to accurately predict real-world events, but he does see their study as a challenge to economists to discontinue their current assumptions about predicting social behavior, like setting interest rates or merchants choosing suppliers.
Perspective Shift
When studying various research fields, from the Earth and Universe to consciousness and Life, we often exclude ourselves. Our habit of setting ourselves apart from everything around us makes us forget that we are just as susceptible to our research results.
We like to think of ourselves as rational decision-makers simply because we possess the ability, but this research is just the latest study to remind us that we are also emotional and often impulsive animals. Some may view this as a bad thing, but I think it’s a doorway opening to a future where we acknowledge all aspects of ourselves and find ways to unite and utilize them. Emotions are equally important as logic, and it’s time we stop assuming we’re either rational or emotional and instead utilize both.
You’re currently reading my free newsletter, Curious Adventure. If you want more, you’ll likely enjoy my other newsletter, Curious Life, which you’ve already received sneak peeks of on Monday mornings. The subscription helps me pay my bills so I can continue doing what I love — following my curiosities and sharing what I learn with you.
Also, if you enjoy my work and want to show me support, you can donate to my PalPal. Thank you for reading. I appreciate you.
Thought-provoking article. Rationality and common sense seem to be losing ground to emotion, especially fear, these days. Understanding how both can serve us is optimal, and your article furthers that end.