Mental models are descriptions of patterns in our world. Action models, in contrast, are directive of how to act.

This page is constantly evolving. Most of these models are popularized already. Unique models to a given source will be given a reference when possible. Models with an asterisk (*) imply either my personal instantiation as models, or my unique naming convention for those cases when they have been outlined in similar formulations by others.

Find Match Quality

You might be trained in piano from a young age and specialize in just that. However, your upper bound as a piano player might be the 80th percentile while your potential upper bound in another field—maybe violin, painting, poetry, or physics—may be the 99th percentile. Match quality is the degree to which people or other resources are properly allocated to the use for which they are best fit. As a guide to life, the model of match quality implies that generalizing early gives more ability to find fields where one can specialize later.

Related: Comparative Advantage

Stack Talents

It is difficult to be the top one percent in the world in any single domain, but much easier in a specific set of domains. If you can find two orthogonal fields where you are in the top ten percent, then you are likely to be in the top one percent of people in the combination of those two fields (10% × 10% = 1%). Top twenty percent in three domains would also put you in the top one percent at the combination of those three. Find where the intersection of your skills makes you uniquely talented for some role, and leverage this towards personal success. Scott Adams, creator of Dilbert, recognized that he was not the funniest person he knew, nor the best artist, but among the top funny artists he knew, which is how he carved himself a spot among the legendary cartoonists.

Reference: Scott Adams’s How to Fail at Almost Everything and Still Win Big, and his blog

Use Situational Reframing

When we think, we are operating under assumptions. Situational reframing can be the attempt to view a situation from another perspective or under different assumptions. For example, you might be operating under the assumption that it’s impossible to accomplish some goal under some constraints. Reframe by removing those assumptions, and then questioning how one might act.

“If you have a 10-year plan of how to get [somewhere], you should ask: Why can’t you do this in 6 months?”

Peter Thiel, Founder of PayPal

Use Granny’s Rule

“Eat your vegetables if you want desert.” As a matter of habit stacking, commit to performing less-desired actions before more-desired ones. Commit to exercising before you watch television, or practicing piano before video games.

Reference: Charlie Munger, Poor Charlie’s Almanack

Steering Conversation

Restate the Opponent’s Argument

If we value in conversation that interlocutors understand the meaning another intends to convey, then one should be able to restate an opponent’s argument in a way their opponent can agree with. Failing to do so would imply that one misunderstands what the other means. If you attack an argument that your opponent does not actually hold or agree with, then you are engaging in the strawman fallacy.

Steelman

A stronger version of restating the opponents argument would be able to recognize what they mean, and depicting them in the most lucid possible way. This is the first of Rapoport’s Rules of conversation. As philosopher Dan Dennett summarizes it: “You should attempt to re-express your target’s position so clearly, vividly, and fairly that your target says, ‘Thanks, I wish I’d thought of putting it that way.'” This is the opposite of the strawman argument, which is restating or misstating your opponents views in a weaker way. When one attacks a steelman and wins, they demonstrate a stronger argument.

Model Language

Language Modeling is the technique of procuring behavior from an interlocutor by demonstrating it yourself. This technique can help direct conversation in productive directions. From How to Have Impossible Conversations: “Model the behavior you want to see in your conversation partner. If you want her to answer a direct question, answer a direct question. If you want her to be patient and listen, be patient and listen. If you want her to start screaming, start screaming. If you want your partner to be open to changing her mind, be open to changing yours. If you want them to be civil, be civil. If you want them to give ground, give some of yours.”

Related: The Chameleon Effect

Altercast

Altercasting is the technique of procuring behavior from an interlocutor by ascribing that behavior as a part of their identity. From How to Have Impossible Conversations: “If you say to someone who’s texting, ‘Wow, you’re a really fast texter,’ you’ve altercasted them as a fast texter. They’ll then embrace that role and want to text more quickly… Altercast your partner into the role of better conversationalist. Say, ‘You’re good at having civil conversations.’ Or, simply, ‘You’re good at keeping your cool.'”

Reference: James Lindsay and Peter Boghossian, How to Have Impossible Conversations

Use Scales

Attempting to quantify belief to a scale can be a useful conversational technique. Consider questions like the following: “On a scale of 1-10, how sure are you in such a belief? As high as a 9, what is giving you doubts? What evidence would have to exist to give to make you less confident? (See counterfactuals)”

Reference: James Lindsay and Peter Boghossian, How to Have Impossible Conversations

Question Epistemology

Epistemology is the process by which holds a belief. If your conversation partner holds a belief particularly dear, question how they came to such a belief. Consider questions like “Is the process by which you came to believe faultless? Is everything your parents think, or everything that those around you think sure to be true? Then how can you be so sure of this belief?”

Reference: James Lindsay and Peter Boghossian, How to Have Impossible Conversations

Appeal to Superordinate Identities

Superordinate identities are those that are inclusive of many sub-identities. From How to Have Impossible Conversations: “When a conversation centers on race, gender, or any other divisive marker in identity politics, people can become defensive and tempers can flare. If you find the conversation getting heated or stuck, shift the focus to superordinate identity markers instead. Rather than dividing, these unify people… Crudely, “You’re white (or Muslim) and I’m black (or Christian), but so what, because we’re both Americans and both human beings.” Notice how this statement moves the conversation toward common ground at the identity level.

Reference: James Lindsay and Peter Boghossian, How to Have Impossible Conversations

Potential

Consider an end worth maximizing, be it profit, well-being, or otherwise. The maximum possible realization of these ends by a business, sports team, person, or other subject can be considered it’s full potential, or an ideal. Indecision, mistakes, and other realities mean we most often fail to reach or full potential. The existence of ideals automatically implies judgment as we fall short of them. Businesses and CEOs can be judged not by how well they are doing, but how well they are doing compared to how well they could be doing. Individuals looking to improve themselves might not consider how good and righteous they are, but consider how good they could be. Failing to reach ideals means there is room for improvement, by definition.

Ideal Divergence*

If you play the AlphaZero chess engine, you will lose. Not because AlphaZero is unbeatable, but because you will make more inaccuracies, mistakes, and blunders. One can reframe the process behind deciding on the right move instead as failing to make the wrong move.

Normally, we look to explain why people think certain ways. Why did that sports agent choose a certain player? Why did that broker prefer a certain stock? This is the bottom-up approach. We can reframe the consideration of why certain decisions were made in a new way: why were the ideal decisions not made? Approach the problem from the viewpoint of the perfect outcome, and look back to see how decision-makers may be misled.

Pre-Mortem

A localized version of Divergence from Ideals to a specific problem. Post-mortem’s are the process of looking for the cause of death after it happened. A pre-mortem is anticipating causes of death prior to execution, so they may be avoided. Before launching a new product, executing a military mission, starting a new initiative, or any other task at hand, imagine your plan has already failed. Ask, why did we fail? What are the causal factors that can be avoided?

Question the Norm

Why is the standard the way it is? Could what’s normal have turned out as something different in different circumstances? Take the American cultural practice of circumcising its male infants. Could you imagine us starting this practice from scratch today? It seems a historical legacy which we continue either by tradition or social proof. We might try to backwards-justify the practice by claiming there are health benefits, but if we imagine thinking from first principles we might not expect the evidence to give sufficiently support the practice.