meta-mental-models
用多學科的心智模型從不同角度看同一個決策
中文翻譯
---
name: "meta-mental-models"
description: "運用來自多個學科的心智模型格狀架構（latticework）來提升決策品質。使用者需要想得更清楚、避開認知盲點、運用跨學科推理，或要從多個角度評估複雜決策時使用，就算他們說的是「我該怎麼想這件事」、「我漏了什麼」、「給我一個不同的視角」或「這裡適用哪些框架」也一樣。"
metadata:
category: "WP-22 跨學科"
tags: ["meta-thinking", "mental-models", "munger", "decision-making"]
---
# 心智模型工具箱
## 框架
```
IRON LAW: Use Multiple Models, Not Just Your Favorite
"To a man with a hammer, everything looks like a nail." (Munger)
A single mental model creates blind spots. Apply 2-3 models from
DIFFERENT disciplines to any important decision. Where models agree,
confidence is high. Where they disagree, the disagreement reveals
the most important dimension of the decision.
```
### 核心心智模型（跨學科）
**來自物理／工程**
| 模型 | 原則 | 應用 |
|-------|-----------|------------|
| **逆向思考（Inversion）** | 不要問「我要怎麼成功？」，改問「我會怎麼失敗？」然後避開它。 | 風險管理、pre-mortem |
| **二階效應（Second-order effects）** | 每個行動都有後果，而那些後果又有後果。要想兩步以後。 | 政策設計、策略 |
| **熵（Entropy）** | 系統在沒有能量投入時會趨向無序。事物預設就會衰敗。 | 維護、品質、人際關係 |
**來自生物學**
| 模型 | 原則 | 應用 |
|-------|-----------|------------|
| **演化／自然選擇** | 存活下來的是適應的，不是絕對意義上「最好」的。 | 市場競爭、產品市場適配（product-market fit） |
| **紅皇后效應（Red Queen effect）** | 你得不斷進步，才能留在原地（因為競爭對手也在進步）。 | 競爭策略 |
| **利基專業化（Niche specialization）** | 通才和專才並存，因為他們服務的是不同的利基。 | 市場定位、職涯策略 |
**來自數學／統計**
| 模型 | 原則 | 應用 |
|-------|-----------|------------|
| **柏拉圖法則（80/20）** | 約 80% 的效果來自約 20% 的原因。 | 排優先順序、資源分配 |
| **均值回歸（Regression to the mean）** | 極端的結果之後，往往跟著比較平均的結果。 | 績效評估、預測 |
| **貝氏定理（Bayes' theorem）** | 根據新證據更新信念，並以先驗機率加權。 | 不確定情況下的決策 |
**來自心理學**
| 模型 | 原則 | 應用 |
|-------|-----------|------------|
| **誘因造成的偏誤（Incentive-caused bias）** | 人會做被誘因驅動的事，不是你要求他們做的事。 | 薪酬設計、政策設計 |
| **能力圈（Circle of competence）** | 知道自己懂什麼、不懂什麼。高風險的決定留在自己的專業範圍內。 | 自我覺察、授權 |
| **漢隆剃刀（Hanlon's razor）** | 能用無知或無能充分解釋的事，就不要歸咎於惡意。 | 衝突化解、職場動態 |
### 應用方法
1. **說明決策或問題**
2. **選 2-3 個相關的模型**，來自不同學科
3. **把每個模型套用到這個情境**：它暗示了什麼？
4. **比較結論**：哪裡一致？哪裡不一致？
5. **綜合**：不一致的地方，就是要解決的關鍵取捨
## 輸出格式
```markdown
# Multi-Model Analysis: {Decision}
## Models Applied
| Model | Discipline | Insight |
|-------|-----------|---------|
| {model 1} | {field} | {what this model says about the situation} |
| {model 2} | {field} | {what this model says} |
| {model 3} | {field} | {what this model says} |
## Convergence
{Where models agree — high confidence}
## Divergence
{Where models disagree — key trade-off to resolve}
## Synthesis
{Recommended decision based on multi-model analysis}
```
## 注意事項
- **模型是簡化**：每個模型都會漏掉一些東西。地圖不等於疆域。把模型當成鏡片，不要當成真理。
- **模型庫會隨時間成長**：從 10-15 個核心模型開始。碰到新領域再加。應用的品質比模型的數量重要。
- **有些模型天生就會衝突**：逆向思考說「避開失敗」。演化說「失敗是你學習的方式」。衝突要靠情境來解決：避開災難性的失敗，擁抱可復原的失敗。
- **不要硬套**：不是每個模型都適用每個情境。如果某個模型沒辦法自然地照亮這個問題，就跳過，不要硬拗來套。
## 參考資料
- 擴充版的心智模型目錄（50 個以上）請看 `references/mental-models-catalog.md`英文原文
---
name: "meta-mental-models"
description: "Apply a latticework of mental models from multiple disciplines to improve decision quality. Use this skill when the user needs to think more clearly, avoid cognitive blind spots, apply cross-disciplinary reasoning, or evaluate a complex decision from multiple angles — even if they say 'how should I think about this', 'what am I missing', 'give me a different perspective', or 'what frameworks apply here'."
metadata:
category: "WP-22 跨學科"
tags: ["meta-thinking", "mental-models", "munger", "decision-making"]
---
# Mental Models Toolkit
## Framework
```
IRON LAW: Use Multiple Models, Not Just Your Favorite
"To a man with a hammer, everything looks like a nail." (Munger)
A single mental model creates blind spots. Apply 2-3 models from
DIFFERENT disciplines to any important decision. Where models agree,
confidence is high. Where they disagree, the disagreement reveals
the most important dimension of the decision.
```
### Core Mental Models (Cross-Disciplinary)
**From Physics/Engineering**
| Model | Principle | Application |
|-------|-----------|------------|
| **Inversion** | Instead of "how do I succeed?", ask "how would I fail?" Then avoid that. | Risk management, pre-mortem |
| **Second-order effects** | Every action has consequences, which have consequences. Think two steps ahead. | Policy design, strategy |
| **Entropy** | Systems tend toward disorder without energy input. Things decay by default. | Maintenance, quality, relationships |
**From Biology**
| Model | Principle | Application |
|-------|-----------|------------|
| **Evolution/natural selection** | What survives is what's adapted, not what's "best" in absolute terms. | Market competition, product-market fit |
| **Red Queen effect** | You must keep improving just to stay in the same place (because competitors improve too). | Competitive strategy |
| **Niche specialization** | Generalists and specialists coexist because they serve different niches. | Market positioning, career strategy |
**From Mathematics/Statistics**
| Model | Principle | Application |
|-------|-----------|------------|
| **Pareto principle (80/20)** | ~80% of effects come from ~20% of causes. | Prioritization, resource allocation |
| **Regression to the mean** | Extreme results tend to be followed by more average ones. | Performance evaluation, forecasting |
| **Bayes' theorem** | Update beliefs based on new evidence, weighted by prior probability. | Decision-making under uncertainty |
**From Psychology**
| Model | Principle | Application |
|-------|-----------|------------|
| **Incentive-caused bias** | People do what they're incentivized to do, not what you ask them to do. | Compensation design, policy design |
| **Circle of competence** | Know what you know and what you don't. Stay within your expertise for high-stakes decisions. | Self-awareness, delegation |
| **Hanlon's razor** | Never attribute to malice what is adequately explained by ignorance or incompetence. | Conflict resolution, workplace dynamics |
### Application Method
1. **State the decision or problem**
2. **Select 2-3 relevant models** from different disciplines
3. **Apply each model** to the situation — what does it suggest?
4. **Compare conclusions** — where do they agree? Where do they disagree?
5. **Synthesize** — the disagreement reveals the key trade-off to resolve
## Output Format
```markdown
# Multi-Model Analysis: {Decision}
## Models Applied
| Model | Discipline | Insight |
|-------|-----------|---------|
| {model 1} | {field} | {what this model says about the situation} |
| {model 2} | {field} | {what this model says} |
| {model 3} | {field} | {what this model says} |
## Convergence
{Where models agree — high confidence}
## Divergence
{Where models disagree — key trade-off to resolve}
## Synthesis
{Recommended decision based on multi-model analysis}
```
## Gotchas
- **Models are simplifications**: Every model omits something. The map is not the territory. Use models as lenses, not as truth.
- **Model inventory grows over time**: Start with 10-15 core models. Add new ones as you encounter new domains. Quality of application matters more than quantity of models.
- **Some models conflict by design**: Inversion says "avoid failure." Evolution says "failure is how you learn." The conflict is resolved by context: avoid catastrophic failure, embrace recoverable failure.
- **Don't force-fit**: Not every model applies to every situation. If a model doesn't naturally illuminate the problem, skip it — don't stretch it to fit.
## References
- For expanded mental models catalog (50+), see `references/mental-models-catalog.md`