멘탈 모델 연구의 공통점과 차이점

<멘탈 모델 연구의 공통점과 차이점>

이석재 Ph.D., PCC, CPCC

멘탈 모델링 코칭 창시자

2026. 5. 16

1. 메타 AI에게 이번에 출간한 <멘탈 모델링 코칭>의 내용과 생각 파트너의 블로그 등을 소개하면서 멘탈 모델 연구의 주요 학자인 Craik, Argyris/Schön, Munger/Parrish, Lee & Lee 2026를 비교분석하도록 요청했습니다. 이 주제에 관심 있는 분들의 방문을 환영합니다.

 

2. 메타 AI가 <멘탈 모델링 코칭>에 대해 전혀 아는 모르더군요. 이와 관련된 생각 파트너의 코칭 활동과 연구 결과물, 책 등을 참고자료로 소개했습니다. 생각 파트너의 코칭 연구를 국외 인물에게 소개하기 위해 영문으로 소통했습니다.

 

How Lee & Lee relate to the broader mental-models tradition 

Kenneth Craik, 1943

Origin of the term.

Craik proposed humans carry a small-scale internal model of external reality and use it to try out alternatives mentally. It was a cognitive-science hypothesis about representation and prediction.

Link: Lee & Lee keep Craik’s idea that we live by an internal model, but move it from descriptive cognitive theory to prescriptive coaching intervention.

Difference: Craik is about how the mind works; Lee & Lee are about how to deliberately diagnose and rebuild that working model for optimal functioning.

Chris Argyris & Donald Schön, 1970s-1990s

Organizational learning and action science.

Mental models = deeply held internal images of how the world works that govern action. Argyris distinguished espoused theory vs theory-in-use, and single-loop vs double-loop learning. Change requires surfacing and testing governing variables.

Link: Lee & Lee’s “Courage to Unlearn” and “Deconstructing the Familiar Self” is double-loop learning in coaching language. The ABC Framework mirrors Argyris’ A → B → C causal chain.

Difference: Argyris is organizational and defensive-routine focused. Lee & Lee add a neuroscience framing, a structured 3S-FORM engine, and an explicit internalization + self-coaching phase for individual sustainability. They also scale to WE via MEWEMIND, which is more ontological-relational than Argyris’ behavioral science.

Charlie Munger & the Farnam Street tradition

Practical decision-making latticework.

Munger popularized collecting mental models from multiple disciplines - inversion, second-order thinking, opportunity cost, compounding, etc.- and using them as a checklist for better judgment. Shane Parrish emphasizes building a latticework and reflecting to refine models.

Link: Both traditions value models as tools for better outcomes. Lee & Lee’s “Behavioral Experiments for New Restructuring” echoes Parrish’s learning loop: experience → reflection → model refinement.

Difference in scope:

• Munger/Parrish: Epistemic toolbox. You acquire and apply many external models to think more clearly.
• Lee & Lee: Ontological restructuring. You work on your core generative model of self-world, via Diagnosis-Change-Execution, with 3S as inner engine and FORM as process scaffold.

Munger asks “What model from biology/economics fits this problem?” Lee & Lee ask “What model is running me, and how do I rebuild it so optimal functioning emerges automatically?”

Difference in method:

• Munger: intellectual breadth + inversion + checklists.
• Lee & Lee: coaching alliance + unlearning + cognitive error restructuring + behavioral experiments + internalization + organizational MEWEMIND diffusion.

Summary comparison

Dimension	Craik	Argyris/Schön	Munger/Parrish	Lee & Lee 2026
Primary goal	Explain cognition	Improve organizational learning	Improve judgment & decisions	Restructure being for optimal functioning
Unit of change	Internal representation	Theories-in-use & defensive routines	Individual thinker’s toolkit	Embedded being in open system
Key mechanism	Simulation	Double-loop learning	Latticework + reflection	3S-FORM double engine
Social scale	Individual	Team/organization	Individual	ME → WE via MEWEMIND
Success metric	Predictive accuracy	Actionable knowledge	Better decisions	Sustainable optimal functioning + self-coaching

 

Where Lee & Lee add something distinct

1. Neuroscience + coaching integration: They explicitly frame mental modeling as neural network restructuring, not just perspective shift.
2. 3S-FORM as dual engine: Inner work 3S and process scaffold FORM are paired, which is more operationalized than Argyris’ inquiry skills or Munger’s checklists.
3. MEWEMIND ontology: The ME↔WE inversion makes relational growth constitutive, not additive. Change is not complete until it propagates to work, relationships and organization.
4. Self-coach as endpoint: The final chapter aims to make the client their own reliable partner, closing the loop on dependency.

Reference

 

이종서 · 이석재 (2026). 멘탈 모델링 코칭. 서울: 코칭북스.

출처: https://coachall.tistory.com/1648 [코치올:티스토리]