100 Mental Models Everyone Should Know and Use
- beyondhorizon965
- Aug 8
- 9 min read
Article
Good thinking is not about having a high IQ. It is about using better tools.
Mental models are those tools. They are the reusable ideas, patterns, and frameworks that help you understand how the world works. A good model helps you ask better questions, avoid predictable mistakes, and make decisions with less noise.
No single model is enough. The point is to build a “latticework”, a term often associated with Charlie Munger, where ideas from psychology, physics, economics, biology, maths, and history strengthen each other.
This guide gives you 100 mental models everyone should know and use, with plain-English meanings and practical examples.
How to use mental models without fooling yourself
Mental models are useful, but they are not magic words. The common mistake is to learn the name of a model and then force it onto every problem.
Use them like lenses, not rules.
Ask:
What does this model reveal?
What does it hide?
What would change my mind?
Which other model disagrees with this one?
A model is strongest when it helps you predict, decide, or explain. If it only sounds clever, leave it out.
The 100 mental models
Better decision-making models
No. | Mental model | What it means | How to use it |
1 | First principles thinking | Break a problem down to basic truths. | Ask what must be true before accepting assumptions. |
2 | Second-order thinking | Consider the results of the results. | Before choosing, ask what happens next. |
3 | Inversion | Solve backwards by asking what would cause failure. | List how to ruin the outcome, then avoid those actions. |
4 | Opportunity cost | Every choice costs the next best alternative. | Compare options against what you give up. |
5 | Trade-offs | Most decisions exchange one good thing for another. | Name the cost clearly before committing. |
6 | Satisficing | Choose good enough when perfect is too costly. | Use for low-risk decisions with limited upside. |
7 | Expected value | Weigh outcomes by probability and payoff. | Useful for bets, investments, hiring, and risk. |
8 | Margin of safety | Leave room for error. | Add buffers to budgets, schedules, and estimates. |
9 | Reversibility | Some decisions are easy to undo, others are not. | Move fast on reversible choices, slow down on irreversible ones. |
10 | Regret minimisation | Choose the option your future self is least likely to regret. | Helpful when values matter more than numbers. |
Models for clear thinking
No. | Mental model | What it means | How to use it |
11 | Occam’s razor | Prefer the simpler explanation when evidence is equal. | Do not add complexity without need. |
12 | Hanlon’s razor | Do not assume malice when error explains behaviour. | Reduces anger and improves judgement. |
13 | Map and territory | A model of reality is not reality itself. | Check whether your data matches the world. |
14 | Circle of competence | Know what you understand and what you do not. | Stay humble outside your expertise. |
15 | Confirmation bias | People favour evidence that supports existing beliefs. | Seek disconfirming evidence on purpose. |
16 | Availability heuristic | Vivid examples feel more common than they are. | Ask for base rates, not just stories. |
17 | Anchoring | The first number or idea shapes later judgement. | Set your own estimate before seeing others. |
18 | Framing effect | The way information is presented changes decisions. | Reword the problem and compare reactions. |
19 | Survivorship bias | Winners are visible, failures are often hidden. | Study the people and projects that failed too. |
20 | Fundamental attribution error | We blame character for others and circumstances for ourselves. | Ask what situation might explain behaviour. |
Models from maths and probability
No. | Mental model | What it means | How to use it |
21 | Base rates | General averages matter before specific details. | Start with the typical outcome. |
22 | Bayesian updating | Change your belief as new evidence arrives. | Treat confidence as adjustable. |
23 | Regression to the mean | Extreme outcomes often move back towards average. | Do not overreact to one great or terrible result. |
24 | Power laws | A few inputs can create most results. | Look for uneven distribution. |
25 | Pareto principle | Roughly 80% of results often come from 20% of causes. | Find the vital few. |
26 | Law of large numbers | More samples give more reliable averages. | Avoid strong claims from tiny samples. |
27 | Compounding | Small gains build into large outcomes over time. | Apply to money, knowledge, trust, and health habits. |
28 | Variance | Outcomes can swing widely even when averages look stable. | Prepare for the range, not just the mean. |
29 | Correlation and causation | Things can move together without one causing the other. | Look for mechanisms and experiments. |
30 | Diminishing returns | Extra effort eventually adds less value. | Stop when the next unit is not worth it. |
Models from systems thinking
No. | Mental model | What it means | How to use it |
31 | Feedback loops | Outputs influence future inputs. | Find what reinforces or balances behaviour. |
32 | Bottlenecks | A system is limited by its tightest constraint. | Fix the constraint before improving elsewhere. |
33 | Constraints | Limits shape what is possible. | Design around the real limit, not the visible annoyance. |
34 | Emergence | Simple parts can create complex behaviour. | Watch the whole system, not only individuals. |
35 | Equilibrium | Systems settle into stable patterns. | Ask what keeps the current state in place. |
36 | Path dependence | Past choices shape present options. | Consider switching costs and inherited habits. |
37 | Tipping points | Small changes can trigger large shifts. | Look for thresholds, not just steady progress. |
38 | Network effects | A product or system gains value as more people use it. | Useful for marketplaces, languages, and communities. |
39 | Tragedy of the commons | Shared resources can be overused by individual incentives. | Align personal benefit with group health. |
40 | Leverage points | Small changes in the right place create large effects. | Change incentives, rules, or goals, not just symptoms. |
Models from economics and incentives
No. | Mental model | What it means | How to use it |
41 | Incentives | People respond to rewards and penalties. | Ask what behaviour the system pays for. |
42 | Principal-agent problem | Agents may not act in the principal’s interest. | Align incentives and increase transparency. |
43 | Moral hazard | Protection from consequences can increase risk-taking. | Keep some responsibility with the decision-maker. |
44 | Adverse selection | Hidden information can attract the wrong participants. | Look for signals that separate quality. |
45 | Supply and demand | Prices and behaviour change with scarcity and desire. | Ask which side of the market shifted. |
46 | Scarcity | Limited resources increase value and tension. | Decide what deserves priority. |
47 | Comparative advantage | Do what you are relatively best at. | Divide work by relative strength, not absolute skill. |
48 | Sunk cost fallacy | Past costs should not justify future waste. | Ask whether you would start again today. |
49 | Externalities | Actions create costs or benefits for others. | Account for effects beyond the direct transaction. |
50 | Creative destruction | New methods replace older ones. | Expect progress to create winners and losers. |
Models from psychology and behaviour
No. | Mental model | What it means | How to use it |
51 | Loss aversion | Losses often feel stronger than equal gains. | Watch for fear-based decisions. |
52 | Status quo bias | People prefer the current state. | Ask whether the default is truly best. |
53 | Social proof | People copy what others do. | Use with care, popularity is not proof. |
54 | Reciprocity | People tend to return favours. | Build trust through genuine help. |
55 | Commitment and consistency | People stick with prior positions. | Make small commitments visible. |
56 | Incentive-caused bias | Rewards distort judgement. | Ask how someone benefits from being right. |
57 | Overconfidence effect | People overrate their accuracy. | Use ranges and track predictions. |
58 | Dunning-Kruger effect | Low skill can hide poor self-assessment. | Seek feedback before confidence. |
59 | Hindsight bias | Past events look obvious after they happen. | Keep decision journals. |
60 | Identity protective cognition | People defend beliefs tied to identity. | Separate ideas from self-worth. |
Models from biology and evolution
No. | Mental model | What it means | How to use it |
61 | Adaptation | Organisms and systems adjust to conditions. | Expect behaviour to fit the environment. |
62 | Natural selection | Traits that improve survival spread. | Ask what the environment rewards. |
63 | Redundancy | Backup systems increase survival. | Build spare capacity for important things. |
64 | Ecosystems | Organisms rely on interdependent relationships. | Study connections, not isolated parts. |
65 | Niches | Success often comes from serving a specific role. | Find where your strengths fit best. |
66 | Homeostasis | Living systems resist change to stay stable. | Expect pushback when changing habits. |
67 | Signalling | Traits or actions communicate hidden information. | Ask what a signal costs to fake. |
68 | Co-evolution | Species, tools, and behaviours change together. | Expect competitors and users to adapt. |
69 | The red queen effect | You may need to improve just to stay in place. | Keep learning in competitive fields. |
70 | Hormesis | Small stress can strengthen a system. | Use manageable challenge, not overload. |
Models from physics and engineering
No. | Mental model | What it means | How to use it |
71 | Inertia | Objects and habits resist change. | Make starting easier than staying still. |
72 | Entropy | Disorder increases without maintenance. | Build routines for upkeep. |
73 | Friction | Resistance slows movement. | Remove small barriers to desired behaviour. |
74 | Critical mass | Enough quantity changes the system’s behaviour. | Build to the point where momentum takes over. |
75 | Load-bearing capacity | Systems fail when stress exceeds design. | Know limits before adding pressure. |
76 | Fail-safe design | Systems should fail in safe ways. | Plan what happens when things break. |
77 | Redundancy engineering | Important systems need backups. | Avoid single points of failure. |
78 | Modularity | Parts can be changed without breaking the whole. | Build projects in separable pieces. |
79 | Scale effects | What works small may fail large. | Test assumptions at the new size. |
80 | The weakest link | Strength depends on the frailest part. | Improve the failure point first. |
Models for learning and work
No. | Mental model | What it means | How to use it |
81 | Deliberate practice | Improvement needs focused work and feedback. | Practise specific weak points. |
82 | Spaced repetition | Memory improves when review is spread out. | Review before forgetting. |
83 | Interleaving | Mixing related skills improves learning. | Practise varied problems, not just one type. |
84 | Feynman technique | Explain simply to reveal gaps. | Teach the idea in plain words. |
85 | Chunking | Group information into meaningful units. | Turn complexity into patterns. |
86 | Flow | Deep focus happens when challenge matches skill. | Reduce distraction and set clear goals. |
87 | Parkinson’s law | Work expands to fill available time. | Set shorter, real deadlines. |
88 | Deep work | Hard thinking needs uninterrupted time. | Protect blocks for demanding tasks. |
89 | Lindy effect | Some things last longer because they have already lasted. | Respect old ideas that still work. |
90 | Beginner’s mind | Fresh eyes notice what experts overlook. | Ask simple questions without shame. |
Models for strategy, risk, and life
No. | Mental model | What it means | How to use it |
91 | Optionality | More good options increase resilience. | Choose paths that keep future doors open. |
92 | Barbell strategy | Combine extreme safety with selective risk. | Protect the downside while allowing upside. |
93 | Asymmetry | Some choices have more upside than downside. | Seek small losses with large possible gains. |
94 | Skin in the game | People judge better when they share consequences. | Trust advice from those who bear risk. |
95 | Pre-mortem | Imagine failure before it happens. | Ask why the plan failed, then fix weak spots. |
96 | OODA loop | Observe, orient, decide, act. | Move through learning cycles faster. |
97 | Game theory | Outcomes depend on other players’ choices. | Think about incentives and likely responses. |
98 | Reputation | Trust compounds through repeated behaviour. | Treat credibility as a long-term asset. |
99 | Anti-fragility | Some systems gain from disorder. | Build habits and portfolios that benefit from stress. |
100 | Long-term thinking | Time changes what matters. | Choose what still makes sense in ten years. |
How to apply these models in real life
Do not try to memorise all 100 at once. Start with a small set that solves the problems you face most often.
For everyday decisions, use:
Inversion
Opportunity cost
Reversibility
Second-order thinking
Margin of safety
For learning, use:
Deliberate practice
Spaced repetition
Feynman technique
Chunking
Feedback loops
For risk, use:
Expected value
Base rates
Pre-mortem
Barbell strategy
Skin in the game
The simplest practical method is a two-minute checklist:
What is the real problem?
What assumptions am I making?
What would prove me wrong?
What is the cost of being wrong?
Which model gives the clearest next step?
Common mistakes when using mental models
The biggest mistake is model worship. A model is a simplification. It leaves things out by design.
Other traps include:
Overfitting one model
Using incentives to explain all behaviour, or psychology to explain every market, creates blind spots.
Ignoring context
A model that works in investing may not work in parenting, health, or public policy.
Sounding clever instead of being useful
If the model does not improve the decision, skip it.
Forgetting values
Some decisions are not only about efficiency. Fairness, dignity, loyalty, and purpose matter too.
The best way to build your latticework
Read widely. Keep a decision journal. Compare predictions with outcomes. Learn from domains outside your own.
A strong thinker does not carry 100 slogans. They carry a flexible set of tools and know when each one breaks.
If you enjoy long-form conversations about ideas, decision-making, culture, history, and what comes next, explore the Beyond the Horizon Podcast through the official website, YouTube, Spotify, TikTok, Patreon, X, or LinkedIn.
FAQs
What are mental models?
Mental models are simplified ways to understand reality. They help you make sense of problems, predict outcomes, and choose better actions.
Which mental model should I learn first?
Start with inversion. Ask what would cause failure, then avoid those causes. It is simple, fast, and useful in almost any situation.
Are mental models the same as cognitive biases?
No. Cognitive biases are predictable thinking errors. Mental models are thinking tools. Some models help you spot and reduce biases.
How many mental models do I need?
You do not need hundreds. A dozen well-used models can improve your decisions more than 100 memorised names.
Can mental models make decisions for me?
No. They improve judgement, but they do not replace values, experience, evidence, or responsibility.
Sources
Charlie Munger, Poor Charlie’s Almanack
Daniel Kahneman, Thinking, Fast and Slow
Donella H. Meadows, Thinking in Systems
Nassim Nicholas Taleb, Antifragile and The Black Swan
Philip E. Tetlock and Dan Gardner, Superforecasting
Richard H. Thaler and Cass R. Sunstein, Nudge
Herbert A. Simon’s work on bounded rationality and satisficing
Elinor Ostrom’s work on commons governance
Farnam Street, mental models essays and explainers
Richard Feynman’s lectures and explanations on learning and scientific thinking
Comments