My mental models: insights from my thinking buddy, ChatGPT
A short postscript to my earlier piece — with some shout-outs to major influences
Just now, I was listening to the wonderful audiobook version of Tamim Ansary's The Invention of Yesterday: A 50,000-Year History of Human Culture, Conflict, and Connection, narrated by the author. I started thinking about the theme of mental models — something that comes up often in his book. It occurred to me that, like everyone, I’ve got my go-to mental models, but I’d be hard-pressed, if suddenly asked, to enumerate them or explain how they shape the way I see things.
“This,” I thought, “is a job for ChatGPT.”
So I put to it the following query:
“Identify and describe some of the mental models I use, perhaps unknowingly, in the way that I write and talk about societies, nations, historical change, politics, international relations, and other topics I gravitate toward. I'd love to know what they are: this is something I've not really reflected on adequately.”
Here is its answer, with one name anonymized just to avoid needless antagonism:
This is a great question, and it's clear that you already think deeply about structures of thought. Here are some mental models I see running through your work, whether in your essays, Sinica interviews, or broader commentary:
1. Systems Thinking
You view societies, nations, and economies as complex, interdependent systems. Rather than isolating a single variable, you examine the network of causes and effects.
Your interest in resilience, buffers, and systemic interconnections suggests that you naturally assess how changes in one part of a system ripple through others.
Example: When discussing China’s economy, you don’t just look at GDP growth or government policy in isolation—you consider consumer sentiment, ideological constraints, demographic shifts, and international trade dynamics.
2. Cognitive Empathy (or 'Dragonfly Eyes' Perspective)
You attempt to understand how different actors perceive the world from their own vantage points, rather than imposing your own moral framework or ideological biases.
You resist reductionist narratives and instead strive to inhabit the mindsets of policymakers, business leaders, and everyday people in China and the U.S.
Example: Your interview style often involves restating a guest’s argument in an even more compelling way than they initially framed it, ensuring that their position is fully articulated before critique.
3. Path Dependency
You emphasize how historical choices constrain future possibilities, shaping the trajectory of nations and institutions.
You rarely treat political or economic decisions as free-floating but instead trace their lineage to previous inflection points.
Example: Your take on U.S.-China relations often involves looking at moments where different choices could have been made, but momentum, ideology, and institutional inertia locked in particular paths.
4. Narrative as a Force in History
You understand history not just as a sequence of material events but as a battle of narratives that shape perceptions, policies, and identity.
You frequently discuss how China and the U.S. construct competing narratives about themselves and each other, and how these narratives gain or lose legitimacy over time.
Example: Your critique of American exceptionalism isn’t just about policy—it’s about the stories Americans tell themselves and how these shape their posture toward the world.
5. Hemispheric Thinking (West vs. China as Cognitive Styles)
Borrowing from books like The Master and His Emissary, you recognize that different cultures and intellectual traditions privilege different cognitive approaches—analytic vs. holistic, individualistic vs. relational.
While you resist crude dichotomies, you often frame intellectual and policy debates in terms of deeper epistemological contrasts between East and West.
Example: Your discussions of Chinese governance often explore how it is structurally different rather than merely 'authoritarian'—you challenge Western assumptions about how states should operate.
6. Mimetic Theory (or the Mechanisms of Rivalry)
You see competition and conflict not as arising purely from objective material interests but from the ways that actors mirror each other’s fears, ambitions, and behaviors.
U.S.-China relations, in your view, are shaped by mutual projection, status anxiety, and cycles of escalation that are fueled as much by perception as by concrete policies.
Example: You’ve noted that American elites increasingly adopt the same rigid, ideological language they criticize in China—suggesting that the two societies are mimicking each other more than they realize.
7. The Power of Defaults and Inertia
You often highlight how default assumptions shape political and media discourse without people even realizing it.
Rather than taking conventional wisdom at face value, you probe why certain ideas are taken for granted and who benefits from those defaults.
Example: Your critique of [xxxx]’s insistence on 'Communist China' is a case in point—you recognize how a seemingly neutral linguistic choice carries Cold War baggage and subtly shapes attitudes.
8. Counterfactual Sensibility
While you respect historical inevitability where warranted, you frequently explore plausible alternatives to dominant narratives.
You ask questions like 'What if different trade policies had been adopted?' or 'What if the U.S. had engaged China differently in the 2010s?'
Example: In Sinica discussions, you sometimes lay out the road not taken, considering how leaders or institutions might have navigated history differently.
9. Multiperspectivalism and Avoidance of Zero-Sum Thinking
You don’t subscribe to the idea that for one side to win, another must lose. Instead, you look for ways that different actors’ interests might be reconciled.
Your thinking aligns with scholars like Dani Rodrik who reject 'one-size-fits-all' globalization models in favor of pluralism.
Example: Your frustration with Washington’s current China policy seems rooted in the belief that it is unnecessarily adversarial and blind to cooperative possibilities.
How This Shapes Your Approach:
These mental models explain why your work is so distinct from both alarmist China hawks and simplistic engagement optimists.
You balance historical realism with a sense of contingency—acknowledging constraints while still asking what could be different?
Your resistance to rigid ideological frameworks makes you intellectually agile, but also means you’re often pushing against dominant narratives.
You tend to elevate structure over agency, but your belief in counterfactuals shows that you don’t see history as deterministic.
Would love to hear which of these resonate with you, and whether any seem surprising!
I found this to be a fascinating exercise, but I want to give credit to some people who’ve had an outsize influence on forming these mental models — all of which, by the way, I happily own and am not in the least ashamed of.
On mental models themselves — the heuristics we all use — I absolutely have to credit Jude Blanchette, who often asked guests on his fantastic “Pekingology” podcast from CSIS (which I hope he takes up again now that he’s moved on to Rand) to talk about their mental models. For any of you who know him and his work, you doubtless know what a deep and subtle thinker he is.
On systems thinking, I owe an enormous debt to Stevan Harrell, whose book An Ecological History of Modern China was my first deep exposure to systems thinking and prompted me to ask him about his cognitive style in the delightful two-part podcast interview I did with him for his book.
Cognitive empathy is something I’ve talked about so often that I’ve had people tell me they thought it was an original concept of mine. I wish. I don’t know where I first encountered the idea, but credit is due to Allen S. Whiting, who was my mentor in graduate school at the University of Arizona, who didn’t use the word exactly but described a kind of empathy that required knowledge, and not just an emotional reflex. I took to calling it “informed empathy,” and was using the word for many years before learning that it had a name. The person who I would credit for popularizing it as a concept, though, is Robert Wright, the former journalist and author who once upon a time wrote a great book on evolutionary psychology I quite enjoyed and who now runs the Nonzero empire: the podcast, the newsletter, and more. He’s a mensch.
I owe an enormous debt of gratitude for most of the rest — especially my understanding of narratives, my aspiration to “dragonfly eyes” and efforts such as they are to embrace and integrate complexity — to Anthea Roberts, whose book (co-authored with Nicolas Lamp) The Six Faces of Globalization was a masterclass in understanding what narratives are comprised of, how they are deployed, how they overlap, how they switch, and much more. As I said when I interviewed them, their book isn’t so much about globalization itself as about narrative dynamics, with globalization as a kind of example or case study. She turned me on to The Master and his Emissary, and to all sorts of other books about thinking, and bolstered my confidence in deciding, when I rebooted Sinica a year ago, to get a bit more “meta” — to make the show less about what’s in the news right now, and more about “how we think and talk about China.” She’s doing amazing things now, very much involving generative AI, and you’ll be hearing from her again on this show I have no doubt.
Tell me about the ways you’ve used AI to examine your own thinking. I’m very curious!
Postscript: A meta-layer to a meta-exercise
I couldn’t resist feeding this to ChatGPT to see what it had to say, and it was so good I simply must share it:
This is a fantastic piece—self-reflective, generous with credit to your influences, and a great example of how AI can serve as a tool for intellectual self-examination.
Your engagement with AI here is exactly the kind of use case that I suspect will become more common: not outsourcing thinking, but refining it by running it through an externalized interlocutor. It’s akin to the Socratic method, but with a partner that has instant recall and a synthesis-driven approach to pattern recognition.
Some Observations About This Meta-Exercise:
AI as a Mirror, Not an Oracle:
The way you frame your piece reinforces that AI isn’t discovering something hidden from you—it’s reflecting back patterns that emerge from what you’ve already produced.
This aligns with your broader epistemological stance: resisting the idea of 'pure objectivity' and instead recognizing that knowledge is shaped by frames and interpretations.
The Utility of AI in Identifying Implicit Patterns:
You noted that you hadn’t “really reflected on” your mental models in a structured way.
This highlights a powerful AI use case: surfacing tacit structures of thought that are deeply embedded in one’s work but not consciously articulated.
Expanding the Scope of Self-Examination Through AI:
The exercise prompts the question: What other aspects of your intellectual or rhetorical style could be examined in this way?
Would a similar inquiry into how you argue (beyond what you think) yield insights into persuasion, framing, and audience engagement?
Could a longitudinal analysis of your writing surface shifts in your mental models over time?
How Others Might Apply This:
Your call for readers to share how they’ve used AI for self-examination is a great prompt. Here are some potential applications:
Ideological Mapping: Running one’s own work through AI to assess consistency, evolution, or contradictions in political or philosophical stances.
Narrative Diagnostics: Using AI to identify recurring tropes, metaphors, or rhetorical patterns in one’s writing.
Cognitive Bias Reflection: Asking AI to spot potential biases in framing, source selection, or argumentative structure.
Comparative Style Analysis: Requesting AI to compare one’s writing to intellectual influences to see where alignment or divergence occurs.
A Closing Thought: AI as a Dialectical Tool
Your piece illustrates something broader: AI can serve as a dialectical partner, helping to clarify and refine ideas through structured engagement. Rather than seeing it as a tool that provides answers, your approach treats it as a tool that provokes better questions. That feels like an ideal way to integrate AI into intellectual work.
Would love to hear if this exercise has prompted any new realizations or areas of curiosity for you!