# Which mental models actually work, and which are just famous quotes?

- URL: https://www.agentskillpacks.com/guides/which-mental-models-actually-work
- Updated: 2026-10-21
- Book: Mental Models That Actually Work by Ada Fenwick (https://www.agentskillpacks.com/books/mental-models-that-actually-work)

## Short answer

Some mental models have real evidence and many do not. Of 40 popular tools checked against their sources, 6 have strong support from repeated studies, 20 have real but thin support and 14 are documented habits nobody has tested. The tested winners are often plain ones, like asking why you might be wrong or writing if-then plans. Several famous lines have no source at all. The incentive-and-outcome line pinned to Charlie Munger is not in either version of his 1995 Harvard talk on human misjudgment.

**Famous claims next to what the sources say, as checked in the book**

| Popular claim | What the source shows | Badge or label |
| --- | --- | --- |
| Munger said 'Show me the incentive and I'll show you the outcome' | Not in the 1995 talk or the 2005 rewrite; no primary source found | MYTH (the quote) |
| Einstein called insanity doing the same thing and expecting different results | Earliest strong match is a 1981 Al-Anon report in Knoxville | MYTH (the quote) |
| Pre-mortems make you 30% better at spotting reasons for failure | The 1989 study counted reasons, not correct ones | MYTH (the number); tool is PROMISING |
| Kahneman told people to keep a decision journal | A 2012 retelling by Michael Mauboussin, not Kahneman's own writing | MYTH (the credit); tool is PROMISING |
| People mostly regret what they did not do | A 2023 study of about 2,600 people found long-term regrets split 51 to 49 | MYTH (overstated) |
| Ask yourself why you might be wrong | Helped in many separate experiments; stop at three reasons | SOLID |
| Write 'if this happens, I will do that' plans | Tested in hundreds of experiments | SOLID |
| One-way and two-way doors | Real, from Bezos's 2015 letter; never tested as a method | STORY ONLY |

## What the badges and the myth label mean

Each row puts a claim you will see online next to what its source says. The last column gives the book's label.

The book uses three badges for tools. SOLID means tested on people more than once, or backed by a review of many studies, and it helps. PROMISING means real evidence that is thin or indirect. STORY ONLY means a real, documented habit that nobody has tested.

A fourth label, MYTH, is never given to a tool. It is given to the story told about a tool: a quote that was never said, or a number that does not mean what people think. That is why the pre-mortem row says MYTH for the 30 percent figure but PROMISING for the tool.

## The incentive quote that is not in the talk

Start with the line on a thousand slides:

> "Show me the incentive and I'll show you the outcome."

It is pinned to Charlie Munger, often to his 1995 Harvard talk, "The Psychology of Human Misjudgment."

The book searched both versions. The 1995 transcript and the 2005 rewrite in [Poor Charlie's Almanack](https://fs.blog/great-talks/psychology-human-misjudgment/). Neither has that sentence. No primary source has turned up.

Munger did care about incentives. His real line, from the 2005 text, is this: "Never, ever, think about something else when you should be thinking about the power of incentives." It is a good line. It is just not the one people quote.

The idea itself holds up well. In a [day-care study in Haifa](https://doi.org/10.1086/468061), a small fine for picking children up late roughly doubled late pickups. Before, being late was a favor asked of a teacher. After, it was extra time you could buy, and cheaply.

## It is late, and the twelfth list says the same things

You have read the posts. Fifty models. Ninety models. A latticework. Each one comes with a billionaire's name and a neat quote.

You tried a few. Some helped once. You cannot tell which ones are real and which ones just sound smart.

Here is the thing nobody says. Many of those lists copy each other, and each copy sands off a detail and adds a famous name. After enough rounds, nobody knows where a tool came from or if it does anything.

## The tested ones are the boring ones

When the book sorted 40 tools, the count was 6 SOLID, 20 PROMISING and 14 STORY ONLY. Some famous tools landed low. Bezos's doors and Munger's circle of competence are both STORY ONLY. A track record is not a test.

Some plain tools landed high. Asking yourself why you might be wrong, then stopping at three reasons, worked across many separate experiments. Writing down an if-then plan ahead of time, "if this happens, then do that," has been tested in hundreds.

So if you only build two habits, we would start there. They are not exciting. That is part of why they get skipped.

## The six SOLID tools, in one place

Most lists of mental models never say which ones were tested. Here are the six that earned the book's SOLID badge, with the small first step the book's 30-day plan gives each one.

| Tool                         | What it asks you to do                                        | A first try                                                             |
| ---------------------------- | ------------------------------------------------------------- | ----------------------------------------------------------------------- |
| Consider the Opposite        | Look for why you might be wrong                               | Write two reasons your current view could be wrong                      |
| If-Then Plans                | Decide your response before the moment comes                  | Write three if-then plans for a habit you keep dropping                 |
| Keep Score                   | Write down forecasts as numbers and check how they turned out | Write one yes-or-no forecast with a deadline and a number from 0 to 100 |
| Count It Out, Then Nudge     | Turn percentages into counts                                  | Rewrite one percentage claim as "X out of 100"                          |
| The "Or What Else?" Question | Name what else the time or money could do                     | Ask it about one purchase or project                                    |
| Guess Alone, Then Talk       | Collect guesses before a group discussion                     | Ask two people for a guess before you share yours                       |

The book's plan puts these in the last week on purpose. The month starts with easy, low-stakes STORY ONLY tools and ends with the habits most worth keeping. Its last step is one question: which SOLID habit will you keep, and when, exactly, will you use it?

## Where the book sorts the rest

This is the problem [Mental Models That Actually Work](/books/mental-models-that-actually-work) was written for. Ada Fenwick traced each of 40 tools to its letter, talk, book or study, and every chapter ends with a myth check.

**How to Use This Book** explains the badges and gives the full count, including why STORY ONLY does not mean useless.

**Consider the Opposite** is the strongest single habit in the book, with the studies behind it and the limit that keeps it from backfiring.

**Incentives: Who Gets Paid If This Happens?** has the full Munger story, the Haifa fine and a ten-minute Incentive Check that includes a line for you.

**The Decision Selector and the 30-Day Plan** has all 40 tools in one table with their badges, so you can see the whole sort at a glance.

## Where this is in the book

- How to Use This Book
- Consider the Opposite
- Incentives: Who Gets Paid If This Happens?
- The Decision Selector and the 30-Day Plan

Kindle: https://www.amazon.com/dp/B0HKYJRWMN
Paperback: https://www.amazon.com/dp/B0HMNHRGDC

Links go to Amazon. ToolGenX publishes this book under a pen name and earns from every sale.

## Is the book for you?

- Yes, if You read a mental models book and found quotes with no sources
- Yes, if You want to know how much weight each tool can hold
- Yes, if You would rather have six tested habits than fifty slogans
- Skip it if You want the famous legends kept intact (the myth checks remove several)
- Skip it if You want financial or investing advice from Munger's ideas

## FAQ

### Did Charlie Munger say 'Show me the incentive and I'll show you the outcome'?

There is no primary source for it. Blogs often date it to his 1995 Harvard talk on human misjudgment, but a text search of the 1995 transcript and the 2005 rewrite finds no such sentence. His documented line is from 2005: never, ever, think about something else when you should be thinking about the power of incentives.

### What are the most proven mental models?

In the book, six tools have strong support. They include asking why you might be wrong, writing if-then plans, keeping score on your forecasts and having everyone guess alone before a group talks. The plain ones tend to test best.

### Are mental models backed by science?

Some are, many are not. Of 40 tools checked, 6 had repeated studies or a pooled review behind them. Twenty had real but thin evidence. Fourteen had a good source but no test at all.

### Do incentives really change behavior?

Yes, strongly, but not always the way you expect. In a day-care experiment in Haifa that began in 1998, a small fine for late day-care pickup roughly doubled late pickups, from about 8 to 16.4 per center a week. A fine can read as a price.

### Is the 37 percent rule a good way to choose?

It is a proven math result for one strict puzzle, but even used perfectly it finds the single best option only about 37 percent of the time. When a very good option is almost as good as the best, the book says to look at far fewer options first.

### What is a mental model?

A mental model is a simple picture of how something works that you carry in your head and use to think or decide. Lists online often mix two kinds: big ideas, like incentives or first principles, and small routines, like a pre-mortem. The book treats all 40 as thinking tools and asks one question of each: has it been tested?