If you're going to collect your thoughts on dual-process politics into one place, then I’ll also leave the critiques I left on that video and post in one place. Though a lot of it is already touched on/incorporated, so I will use quotes from the article that touch on the critique before I present it. I will organize it into three main sections/critiques.
1) The Dual-Process Model Still Feels Too Simplistic
You write:
> Before proceeding, a quick disclaimer: while Dual Process Theory is highly influential, it’s not uncontested as a framework for psychological decision-making.
I'm happy you started thinking about critiques of dual-process theory but beyond the problem you linked, I have a some more problems with the theory.
The replication crisis hasn’t been kind to DP, and when you look at it closely, it breaks down in ways that aren’t trivial. Take driving or typing: these are clearly intentional actions, but they’re also not fully conscious. If System 1 is "unconscious and automatic" while System 2 is "conscious and deliberate", then where do these fit? If I’m driving somewhere familiar, I’m not consciously thinking about every turn, but I am intentionally driving to my destination. So is this System 1 or System 2? The distinction just doesn’t hold up well under scrutiny.
Then there’s the issue of unconscious influences on conscious thought. Let’s say I’m actively debating which brand of chocolate to buy, weighing my options logically. But in reality, I’ve already been nudged by past advertising, branding, and placement in the store. Is that System 1 or System 2? If we can’t even cleanly separate the two in simple decisions like this, applying it to political thinking feels even messier. There is a problem with nudged/directed *motivated* reasoning.
This is why I suggested before that it might make more sense to frame political thinking as a spectrum—on one end, people who spend a lot of time and resources analyzing policy; on the other end, people who don’t have that luxury. This framing keeps most of the ideas in your model intact while avoiding the pitfalls of an oversimplified duality.
2) A worry about a label for "System 2 thinker"
You write:
> In practice, most people don’t have the time, energy, or inclination to devote significant mental compute to politics.
I'm happy you incorporated that critique of mine into the post, but I do urge you to go further and examine the material conditions that lead some people to not do it. You touch on it, but, at the risk of sounding like a stereotypical socialist, think about the underlying material conditions and class conflict involved.
I worry that if we accept this dual-process framing uncritically, it might lead to the creation of a self-proclaimed ingroup of "System 2 thinkers": people who see themselves as more rational, more deserving of attention, or more politically sophisticated. The problem becomes pretty obvious when you map it onto the actual political/class/economic landscape.
For example: People love to mock mass movements like BLM for lacking well-formed policy proposals, and using slogans like ACAB instead, but step back for a second—why is that? If you live in an overpoliced neighborhood with lead pipes, food deserts, and constant exposure to environmental hazards, you’re not spending much time writing white papers on policy solutions. This is a class of people too busy trying to survive. Meanwhile, the class of people who do have time and resources to craft detailed policy proposals are, more often than not, the ones who are already doing fine under the status quo.
Political analysis costs time and nutrients. Analysis may be objective, but preferences are subjective. If the group that can write analysis prefers blue buildings while the group that can't prefers red buildings, a world that prioritizes the worldview of "system 2 thinkers" will be filled with blue buildings, even if it's mostly or wholly subjective.
If you have to work two jobs to get by, you’re simply not going to have the energy to sit down and write out a political plan to address your needs and preferences. But that doesn’t mean your grievances aren’t real, or that they don’t matter. The danger of an ingroup of "System 2 thinkers" is that it could just end up reinforcing existing inequalities — because the people who already have the time to think deeply about policy will naturally set the terms of the conversation, while those who don’t will be dismissed as reactionary or uninformed.
3) "Far" and "Extreme" are just code for "Threatening to status-quo"
You write:
"There’s also a danger of misinterpreting what this framework tells us. For example, we shouldn’t assume that radical ideas are wrong by default. Some System 2 superusers end up with radical conclusions (I might have some of my own)."
This is the weakest of my critiques, and may not even apply to this post (so skip this one if you're short on time).
I've touched on before, so I'm happy to see it included. Although our previous discussion on this was about the words "far" and "extreme". Maybe you meant "radical" as similar to "far" in that discussion? In any case the point I’ve brought up before (and I still think applies) is that words like "far" and "radical" and "extreme" aren’t used consistently. They’re not neutral descriptions of how different a belief is from the mainstream; they’re mostly just ways of saying "threatening to power structures."
I used the example of anarcho-primitivism. It’s obviously more "extreme" than, say, a standard environmentalist position, but nobody calls anarcho-primitivists "far-left extremists." Why? Because they’re a bunch of niche philosophers with no real power. They’re radical, but not a threat, so they don’t get branded as extremists. On the other hand, if you advocate for something like worker co-ops at scale, suddenly that’s "radical leftism"—even though it’s way less of a departure from the status quo than the anarcho-primitivists.
But again, this last one is more a word of caution for the language of future posts than a substantial critique.
To address what I see as the most pressing one, I think it would be a grave mistake to dismiss System 1 politics outright. Even a rough, low-pass approximation of reality is still an approximation, and it shouldn’t be ignored. It’s a symptom of something deeper. Those who have the privilege of having the time and energy to System 2 think ought to pay close attention to what System 1 thinkers are saying. Although they might not understand the solution (or the problem in its entirety) it's still a signal that something is off, that some of your fellow humans are suffering. And they are trying to communicate that suffering. It would help to expend compute on how to make things better. What was that thing spyderman's uncle used to say?
On a separate note, your last comment made me realize my last article was kind of banal. You rascal!
The last article? You meant the last one on this topic or your literal last one (about contrarianism)? Your last one on this topic is a great collection of studies; I think even someone who isn't surprised by the argument/conclusion would still greatly appreciate it just for the literature-review-esque nature of it alone.
The last one about contrarianism I liked, though, at the risk of patting myself too much on the back, I think it's improved by the inclusion of my comment, since it then becomes an example of how (in political analysis) the same evidence/dataset can strengthen radically different theories/models.
I do like the article -- but here are some questions to eventually draw you to having the exact same (correct /s) views as myself:
1) What is a process?
2) What does it mean for a human to "process"? ("What are you a Mill? Are you grinding Wheat?" - B.F.Skinner)
3) What does it mean for the dual process theory to "approximate reality"? What does that claim actually mean
Then Extension/Joke Questions:
4) Why was the cognitive revolution a mistake?
5) How can behaviourism/operant conditioning explain mice navigating a certain way around a Maze... (lol: "it definitely can't, so now we need "cognitive processing" -- wait, wtf how does that help)
6) Why only two processes? Why not 3, 7, 21?
7) So is it better to live in a system 2 world where we have 7 hour streams on Joe Rogan about whether the Holocaust actually happened because is there really evidence for any of that stuff, or is it better to have the vast majority of people seeing incredibly concerning politics and saying "fuck you fascist" without the need for endless interminable discussion (against the backdrop of Oxford Oak and Hardback books)
Your comment intrigues me greatly but I'm afraid I will make a fool of myself trying to answer your questions. What do you see as the mistake in dual process theory? And what do you think is the fundamental error behind the cognitive revolution? (interesting)
That said, if I’m reading you correctly, wouldn’t even a behaviorist framework allow for the idea that longer reflection improves task performance? Perhaps one can do away with dual process theory but maintain the essence of the framework even in a behaviorist setting.
> What do you see as the mistake in dual process theory?
The "dual" and the "process" parts ;)
> Fundamental error behind the cognitive revolution?
Lack of empirical content. Metaphysical speculation. Deployment of underanalysed computational metaphors.
> wouldn’t even a behaviorist framework allow for the idea that longer reflection improves task performance?
Sort of, Im also not whole heartedly endorsing 1970's Behaviourism either. However, I think Behaviourism would lend itself better to making clear what we are measuring, and whether that is relevantly analogous to Politics to be informative to anything in that.
> Perhaps one can do away with dual process theory but maintain the essence of the framework even in a behaviorist setting.
On this last point, I don't endorse throwing it away, I think there is value in its use. Just have to be careful on whether we're reading metaphysics off of that and in the limits of our inferences.
I also have a comment that has nothing to do with the post itself, but is rather about the thumbnail. I find it fascinating how generative AI just can't do pixel art.
Long long ago I had a Youtube channel for which I did some pixel art. When I started learning it I came across the adage "pixel art is more sculpting than drawing". This turned out to be an apt metaphor; you must be very selective about which pixels to include or exclude since you have so few to work with and you can't fudge things by making some pixels larger, or placing them in between other pixels. Generative AI doesn't seem to get this and just fudges things the whole time, totally missing the point of what makes pixel art so cool.
People often point to generative AI struggling with high detail art (wrong shadows, wrong light sources etc), but just as interesting is the fact that it struggles with low detail art (like pixel art).
Thank you! Yes, I would consider naive realism as resulting from System 1 thinking, as an efficient first pass approximation of reality. I notice that Dan Williams doesn't consider this among the "roots" he cites (but it seems quite plausible). Thank you for this!
"Could certain political ideologies naturally appeal more to one cognitive style than the other?" Yes.
"...most people who disagree with you likely aren’t mustache-twirling villains—maybe they’ve just processed the same information differently." Yes
"The key is in understanding whether the idea is the result of correct serious reflection or not." Yes, and whether and how it has been tested and evidenced (practice and theory merged).
I too am curious about the obvious relation between thinking styles and politics. I suppose that, in my case, the curiosity amounts to a fascination that began long ago, but that, within the last ten or so years, has become a more structured inquiry. After reading Kahneman and Tversky, Lakoff and Johnson, Bob Altemeyer, and others that have used surveys and observations, I concentrated on readings in neurobiology, as well as works on the evolution of agency and our sociobiology. My main area of interest has been to understand how and why the thinking styles of authoritarian-ideological thinkers differ from those of classical liberals and scientific humanists (see the y-axis of the Political Compass).
I agree that System 1 and System 2 are helpful in distinguishing thinking styles, but I've come to see a spectrum of thinking styles, with various styles (groups of methods) dominating particular people and groups (like habits and personality traits). The physical and social environment will affect what methods dominate. (Stresses that lead to anxiety and fear will shift how one thinks.) Many people will compartmentalize their thinking (the methods they use). An example of compartmentalization are the scientists who decide to think ideologically and metaphysically (dualisms and absolutes) about some questions.
A suggested reading: "The Psychology of Populism: Tribal Challenges to Liberal Democracy", (https://www.cis.org.au/app/uploads/2021/08/op183.pdf ), published by the Centre for Independent Studies (self-described as a classical liberal think tank). It is written from the perspectives of liberal democracy and evolutionary psychology. The author, Joseph P. Forgas, is a Hungarian-Australian social psychologist. I'd really like to know y'all's thoughts on this.
If anyone is interested, I have references from the philosophy of science, evolutionary psychology, neuropsychology, neuropsychiatry, and neurophilosophy. I'd be interested in your references suggestions to further my understanding of this most important topic.
I feel compelled to tell you about Leor Zmigrod's work. I really think you'd like her paper: "A Psychology of Ideology: Unpacking the Psychological Structure of Ideological Thinking."
It would be so nice to discuss this material with people who might critique my notions.
Much of my references and thoughts can be found in my article on Horseshoe theory, I was very interested in the previous article you linked "The Psychology of Populism: Tribal Challenges to Liberal Democracy", seemed really relevant but I'm not sure I have much to add at the moment. Will check out this work by Zmigrod too.
If you're going to collect your thoughts on dual-process politics into one place, then I’ll also leave the critiques I left on that video and post in one place. Though a lot of it is already touched on/incorporated, so I will use quotes from the article that touch on the critique before I present it. I will organize it into three main sections/critiques.
1) The Dual-Process Model Still Feels Too Simplistic
You write:
> Before proceeding, a quick disclaimer: while Dual Process Theory is highly influential, it’s not uncontested as a framework for psychological decision-making.
I'm happy you started thinking about critiques of dual-process theory but beyond the problem you linked, I have a some more problems with the theory.
The replication crisis hasn’t been kind to DP, and when you look at it closely, it breaks down in ways that aren’t trivial. Take driving or typing: these are clearly intentional actions, but they’re also not fully conscious. If System 1 is "unconscious and automatic" while System 2 is "conscious and deliberate", then where do these fit? If I’m driving somewhere familiar, I’m not consciously thinking about every turn, but I am intentionally driving to my destination. So is this System 1 or System 2? The distinction just doesn’t hold up well under scrutiny.
Then there’s the issue of unconscious influences on conscious thought. Let’s say I’m actively debating which brand of chocolate to buy, weighing my options logically. But in reality, I’ve already been nudged by past advertising, branding, and placement in the store. Is that System 1 or System 2? If we can’t even cleanly separate the two in simple decisions like this, applying it to political thinking feels even messier. There is a problem with nudged/directed *motivated* reasoning.
This is why I suggested before that it might make more sense to frame political thinking as a spectrum—on one end, people who spend a lot of time and resources analyzing policy; on the other end, people who don’t have that luxury. This framing keeps most of the ideas in your model intact while avoiding the pitfalls of an oversimplified duality.
2) A worry about a label for "System 2 thinker"
You write:
> In practice, most people don’t have the time, energy, or inclination to devote significant mental compute to politics.
I'm happy you incorporated that critique of mine into the post, but I do urge you to go further and examine the material conditions that lead some people to not do it. You touch on it, but, at the risk of sounding like a stereotypical socialist, think about the underlying material conditions and class conflict involved.
I worry that if we accept this dual-process framing uncritically, it might lead to the creation of a self-proclaimed ingroup of "System 2 thinkers": people who see themselves as more rational, more deserving of attention, or more politically sophisticated. The problem becomes pretty obvious when you map it onto the actual political/class/economic landscape.
For example: People love to mock mass movements like BLM for lacking well-formed policy proposals, and using slogans like ACAB instead, but step back for a second—why is that? If you live in an overpoliced neighborhood with lead pipes, food deserts, and constant exposure to environmental hazards, you’re not spending much time writing white papers on policy solutions. This is a class of people too busy trying to survive. Meanwhile, the class of people who do have time and resources to craft detailed policy proposals are, more often than not, the ones who are already doing fine under the status quo.
Political analysis costs time and nutrients. Analysis may be objective, but preferences are subjective. If the group that can write analysis prefers blue buildings while the group that can't prefers red buildings, a world that prioritizes the worldview of "system 2 thinkers" will be filled with blue buildings, even if it's mostly or wholly subjective.
If you have to work two jobs to get by, you’re simply not going to have the energy to sit down and write out a political plan to address your needs and preferences. But that doesn’t mean your grievances aren’t real, or that they don’t matter. The danger of an ingroup of "System 2 thinkers" is that it could just end up reinforcing existing inequalities — because the people who already have the time to think deeply about policy will naturally set the terms of the conversation, while those who don’t will be dismissed as reactionary or uninformed.
3) "Far" and "Extreme" are just code for "Threatening to status-quo"
You write:
"There’s also a danger of misinterpreting what this framework tells us. For example, we shouldn’t assume that radical ideas are wrong by default. Some System 2 superusers end up with radical conclusions (I might have some of my own)."
This is the weakest of my critiques, and may not even apply to this post (so skip this one if you're short on time).
I've touched on before, so I'm happy to see it included. Although our previous discussion on this was about the words "far" and "extreme". Maybe you meant "radical" as similar to "far" in that discussion? In any case the point I’ve brought up before (and I still think applies) is that words like "far" and "radical" and "extreme" aren’t used consistently. They’re not neutral descriptions of how different a belief is from the mainstream; they’re mostly just ways of saying "threatening to power structures."
I used the example of anarcho-primitivism. It’s obviously more "extreme" than, say, a standard environmentalist position, but nobody calls anarcho-primitivists "far-left extremists." Why? Because they’re a bunch of niche philosophers with no real power. They’re radical, but not a threat, so they don’t get branded as extremists. On the other hand, if you advocate for something like worker co-ops at scale, suddenly that’s "radical leftism"—even though it’s way less of a departure from the status quo than the anarcho-primitivists.
But again, this last one is more a word of caution for the language of future posts than a substantial critique.
Great points.
To address what I see as the most pressing one, I think it would be a grave mistake to dismiss System 1 politics outright. Even a rough, low-pass approximation of reality is still an approximation, and it shouldn’t be ignored. It’s a symptom of something deeper. Those who have the privilege of having the time and energy to System 2 think ought to pay close attention to what System 1 thinkers are saying. Although they might not understand the solution (or the problem in its entirety) it's still a signal that something is off, that some of your fellow humans are suffering. And they are trying to communicate that suffering. It would help to expend compute on how to make things better. What was that thing spyderman's uncle used to say?
On a separate note, your last comment made me realize my last article was kind of banal. You rascal!
The last article? You meant the last one on this topic or your literal last one (about contrarianism)? Your last one on this topic is a great collection of studies; I think even someone who isn't surprised by the argument/conclusion would still greatly appreciate it just for the literature-review-esque nature of it alone.
The last one about contrarianism I liked, though, at the risk of patting myself too much on the back, I think it's improved by the inclusion of my comment, since it then becomes an example of how (in political analysis) the same evidence/dataset can strengthen radically different theories/models.
The one about contrarianism. I don’t know, it’s the same old soup
I don't think anyone could complain that a post with the subtitle "A reminder" includes old arguments.
I do like the article -- but here are some questions to eventually draw you to having the exact same (correct /s) views as myself:
1) What is a process?
2) What does it mean for a human to "process"? ("What are you a Mill? Are you grinding Wheat?" - B.F.Skinner)
3) What does it mean for the dual process theory to "approximate reality"? What does that claim actually mean
Then Extension/Joke Questions:
4) Why was the cognitive revolution a mistake?
5) How can behaviourism/operant conditioning explain mice navigating a certain way around a Maze... (lol: "it definitely can't, so now we need "cognitive processing" -- wait, wtf how does that help)
6) Why only two processes? Why not 3, 7, 21?
7) So is it better to live in a system 2 world where we have 7 hour streams on Joe Rogan about whether the Holocaust actually happened because is there really evidence for any of that stuff, or is it better to have the vast majority of people seeing incredibly concerning politics and saying "fuck you fascist" without the need for endless interminable discussion (against the backdrop of Oxford Oak and Hardback books)
Your comment intrigues me greatly but I'm afraid I will make a fool of myself trying to answer your questions. What do you see as the mistake in dual process theory? And what do you think is the fundamental error behind the cognitive revolution? (interesting)
That said, if I’m reading you correctly, wouldn’t even a behaviorist framework allow for the idea that longer reflection improves task performance? Perhaps one can do away with dual process theory but maintain the essence of the framework even in a behaviorist setting.
> What do you see as the mistake in dual process theory?
The "dual" and the "process" parts ;)
> Fundamental error behind the cognitive revolution?
Lack of empirical content. Metaphysical speculation. Deployment of underanalysed computational metaphors.
> wouldn’t even a behaviorist framework allow for the idea that longer reflection improves task performance?
Sort of, Im also not whole heartedly endorsing 1970's Behaviourism either. However, I think Behaviourism would lend itself better to making clear what we are measuring, and whether that is relevantly analogous to Politics to be informative to anything in that.
> Perhaps one can do away with dual process theory but maintain the essence of the framework even in a behaviorist setting.
On this last point, I don't endorse throwing it away, I think there is value in its use. Just have to be careful on whether we're reading metaphysics off of that and in the limits of our inferences.
> What does it mean for a human to "process"? What are you a Mill?
When Mon0 gave an example of a human with good process, he literally showed you a Mill.
I also have a comment that has nothing to do with the post itself, but is rather about the thumbnail. I find it fascinating how generative AI just can't do pixel art.
Long long ago I had a Youtube channel for which I did some pixel art. When I started learning it I came across the adage "pixel art is more sculpting than drawing". This turned out to be an apt metaphor; you must be very selective about which pixels to include or exclude since you have so few to work with and you can't fudge things by making some pixels larger, or placing them in between other pixels. Generative AI doesn't seem to get this and just fudges things the whole time, totally missing the point of what makes pixel art so cool.
People often point to generative AI struggling with high detail art (wrong shadows, wrong light sources etc), but just as interesting is the fact that it struggles with low detail art (like pixel art).
Great article! What you say about System 1 political thinking is very much like what Dan Williams says about naive realism. https://www.conspicuouscognition.com/p/in-politics-the-truth-is-not-self
Thank you! Yes, I would consider naive realism as resulting from System 1 thinking, as an efficient first pass approximation of reality. I notice that Dan Williams doesn't consider this among the "roots" he cites (but it seems quite plausible). Thank you for this!
Yes, he ought to integrate System 1 thinking into his model of naive realism. Maybe he does somewhere else, as he's written about naive realism a lot.
"Could certain political ideologies naturally appeal more to one cognitive style than the other?" Yes.
"...most people who disagree with you likely aren’t mustache-twirling villains—maybe they’ve just processed the same information differently." Yes
"The key is in understanding whether the idea is the result of correct serious reflection or not." Yes, and whether and how it has been tested and evidenced (practice and theory merged).
I too am curious about the obvious relation between thinking styles and politics. I suppose that, in my case, the curiosity amounts to a fascination that began long ago, but that, within the last ten or so years, has become a more structured inquiry. After reading Kahneman and Tversky, Lakoff and Johnson, Bob Altemeyer, and others that have used surveys and observations, I concentrated on readings in neurobiology, as well as works on the evolution of agency and our sociobiology. My main area of interest has been to understand how and why the thinking styles of authoritarian-ideological thinkers differ from those of classical liberals and scientific humanists (see the y-axis of the Political Compass).
I agree that System 1 and System 2 are helpful in distinguishing thinking styles, but I've come to see a spectrum of thinking styles, with various styles (groups of methods) dominating particular people and groups (like habits and personality traits). The physical and social environment will affect what methods dominate. (Stresses that lead to anxiety and fear will shift how one thinks.) Many people will compartmentalize their thinking (the methods they use). An example of compartmentalization are the scientists who decide to think ideologically and metaphysically (dualisms and absolutes) about some questions.
A suggested reading: "The Psychology of Populism: Tribal Challenges to Liberal Democracy", (https://www.cis.org.au/app/uploads/2021/08/op183.pdf ), published by the Centre for Independent Studies (self-described as a classical liberal think tank). It is written from the perspectives of liberal democracy and evolutionary psychology. The author, Joseph P. Forgas, is a Hungarian-Australian social psychologist. I'd really like to know y'all's thoughts on this.
If anyone is interested, I have references from the philosophy of science, evolutionary psychology, neuropsychology, neuropsychiatry, and neurophilosophy. I'd be interested in your references suggestions to further my understanding of this most important topic.
Thanks a bunch for this!
I feel compelled to tell you about Leor Zmigrod's work. I really think you'd like her paper: "A Psychology of Ideology: Unpacking the Psychological Structure of Ideological Thinking."
It would be so nice to discuss this material with people who might critique my notions.
Much of my references and thoughts can be found in my article on Horseshoe theory, I was very interested in the previous article you linked "The Psychology of Populism: Tribal Challenges to Liberal Democracy", seemed really relevant but I'm not sure I have much to add at the moment. Will check out this work by Zmigrod too.