Why does changing your mind sometimes feel like losing a part of yourself. If beliefs are supposed to track truth, why do they so often track our tribe instead? Why does a challenged belief feel like a personal attack. What does your brain think it's defending when it defends an idea? What if beliefs didn't evolve to be true, but to be useful. What if your strongest convictions are solving social problems, not intellectual ones. Today we'll speak with public intellectual and
neuroscientist Sam Harris about the topic of our beliefs. Welcome to Inner Cosmos with me David Eagleman. I'm a neuroscientist and author at Stanford, and in these episodes we sail deeply into our three pound universe to understand how we see the world and importantly today, what we take to be true about it. Your brain has a lot of beliefs about what is true or false. So if I ask you is Paris the capital of France, your brain immediately judges the truth value of that statement. Is it
accurate or inaccurate? How about if I ask you is ten plus ten twenty one? Your brain has to take in the auditory information and compare it against what it knows and make a decision. And most people would probably agree with your judgments on these questions. But this can of course get more complex when it involves other kinds of beliefs, Like Mohammad was a holy proh fit chosen
by Allah to deliver a message to humanity. Some people think that's a true statement, others think that's false, or the statement early abortion should be supported, or Trump is on balance a good president, or whatever. I'm not suggesting how you should evaluate these sentences. I'm just pointing out there are lots of sentences I can say that you might find true or false. So every moment of your life, your brain is making commitments. You're not necessarily announcing these
or thinking about them consciously. They live under the hood. But these commitments shape what seems true, what feels threatening, what deserves attention, and what can be safely ignored. These commitments we have are what we call beliefs, and beliefs are how the brain turns information into action. In other words, they determine whether an utterance becomes something you respond to or something that fades into the background from the outside.
Beliefs look like opinions, or positions, in other words, something you might argue about or align yourself with. But from the inside they feel fundamental, as in this is true, this is false. Collectively, your beliefs feel like the structure of reality itself. They build the backdrop against which your decisions make sense. Now, the puzzle that's always fascinated me is that intelligent people who are well informed often reach
very different conclusions about their beliefs. So you have evidence that feels decisive to one person and that feels incomplete or unconvincing to another. And what that means is that changing one's mind typically feels really hard, and when you have a disagreement with someone, this can feel personal, like it involves your idea. So when we look at these patterns, we see that belief seems to be doing something more
than building accurate models of the world. That might be part of why the brain builds true and false judgments, but there's probably more involved. For one, beliefs help coordinate your social life by supporting identity within your tribe and establishing alignment between people. Also, they can manage uncertainty by stabilizing emotion. When we look at it through these lenses, belief is, of course a biological process as much as
a philosophical one. Somewhere between encountering a claim and acting on it, your brain has to perform a series of operations that determine whether some statements should be reacted to, or maybe whether it should be questioned or whether it should be dismissed entirely. Whatever your brain decides on the truth value of the statement, that shapes what you do next. And all of this tends to run below the radar
of conscious awareness. Now, understanding what the brain is doing with beliefs is massively important because what you believe to be true and not true determines everything about cooperation or conflict. And so this sits right at the center of all that's happening in politics, in moral judgments, in social trust. So I focus a lot of my episodes on these issues about polarization and in groups and out groups and shibbleeths. But today I want to move upstream a level and
focus on the machinery beneath them. How does the brain decide what to accept as true? How do emotion and identity in social context shape what feels credible? And there's no one better to explore these questions with than Sam Harris, whose work has examined belief from inside the brain and within public discourse. Sam is a public intellectual, a philosopher and author, and also a neuroscientist. He's well known for his writings on rationality, religion, ethics, free will, and meditation.
He has long researched how science can inform human values with books like The End of Faith and The Moral Landscape. He hosts the Making Sense podcast, which focuses on consciousness and society and current events. So today we're going to ask what are beliefs really about. Here's my conversation with Sam Harris. So, Sam, I want to talk with you about belief. Many years ago you did neuroimaging studies on belief,
So let's start there. Tell us how you think about belief and how you measure that in the brain.
Believing things to be true of als versus being uncertain. This is something that our brains do readily, right, This is largely This is what distinguishes us, certainly when we're talking about complex representations of the world. This distinguishes us from our closest primate cousins. I wouldn't say that that other mammals and certainly, you know, the other primates don't have beliefs because clearly they represent states of the world cognitively, and they have expectations on that basis.
Like there'll be grubs under this rock.
Yeah, so they've got, you know, goal seeking behavior, and that those goals are framed by some expectation of the way the world is, and they can be surprised by different outcomes. But we're the only species that were we know of that trades in these linguistic representations of the world. It can be completely divorced from any sensory entanglement with
the state of the world that is being described. I mean, you get a cell phone call and there's a voice on the other end of the line and it says, you know, we've got your daughter and you got to
come up with a million dollars to get her back. Now, if believed all of a sudden, those you know, small mouth noises that came you know through the phone, deletely common deer or you know, physiology, your whole world is all of a sudden about that, you know, kind of bent into the shape of the implications of of that you know utterance. Whereas if it's the wrong number. If you don't, I don't have a daughter, I know what you're talking about, right, just it bounces off, it doesn't,
it doesn't, it doesn't become behaviorally or emotionally actionable. So it was this difference between this fairly gossamer representation of the world that that that it's just a piece of language right in the mind. It could be your own thought, it could be the thought of others you're communicated to you, it could be something you read on on the you know,
the front page of a newspapers. It was it was a transition from that just the mere, mere decoding of a piece of language, to suddenly finding it not only behaviorally actionable, but you know, you know, unthinkable not to
be motivated by those by those you know phonemes. So there's clearly, you know, steps in the chain where you you you you parse language or parse a statement and understand its meaning, and then there's this secondary operation of judging it to be true or false, right, or or just deciding that you can't judge it one or the other. And it seemed to me that on some level, this is the most important thing in the world, right, I mean,
if you're talking about the power of ideas. You first have to stand upstream of that claim to talk about the difference between believing or disbelieving, accepting or rejecting certain ideas right and so, and ideas are obviously everything in our lives. I mean, it's it's how we're making a civilization, is how we're failing to secure a civilization. It's how we're needlessly producing harm for for ourselves and others. I mean, it's just it's it's belief. Is claims to knowledge you
know true or otherwise? Are everything?
Right?
And so the question in my mind is what was the brain doing to find something behaviorally actionable or to find it not based on accepting or rejecting a claim to knowledge about the world.
And so that's binary. But you also looked.
At uncertain certainty as a third condition.
But just so I'm clear, was uncertainty the I just don't know one way or the other.
Yeah, okay, And you clearly, I mean, like you clearly couldn't know. So if you know a question that would provoke that would be to give someone some gigantic number and say that this number is prime. Right, It might be prime ends in a one, it might be prime.
But obviously I don't know, or I have an you know, you have an even number of hairs on your body, right, you know you presumably you don't know, and so you can you can come up with lists of statements that are just as easy to decode as any other statements. But what you come up against, you know, once you understand them, is that there's really no way for you to know what's true.
Got it, So you've got I either believe this to be true some state, give me an example of what you used.
My hypothesis was that belief, disbelief, and uncertainty were content independent operations of the mind and therefore of the brain. So that if I gave you a statement like two plus two equals four versus two plus two equals five, and those are those are parsed and analyzed very similarly. Right, you're just just however, you understand arithmetic at the level of the brain. You're doing that, but you're coming to
the opposite conclusion based on those two examples. But the hypothesis was that if we interrogated many different streams of knowledge and belief formation, so mathematical, logical, autobiographical, religious, political, right, say, the ethical the cognitive and you know, neuro physiological operations required to parse and judge that the truth value of any of those statements had to be very very different.
I mean, what you're doing for arithmetic is not the same as what you're doing for for religion presumably, or ethics, or you know, you know, episodic memories. Say, but there was this final there had to be a bottleneck with which reported acceptance or rejection of the statement or uncertainty.
Great, and so you put people in the scanner and what did you find.
Well, we found that accepting a statement to be true versus rejecting it as false did report a kind of content agnostic you know, subsequent operation. One thing that was interesting, though, is that, and this actually goes back many many centuries to the work of Spinoza. Spinosa hypothesized that merely understanding a statement was akin to accepting it. Like it's like in order to just if I give you, if I give you just any utterance that you you know, any
claim about the world that you can parse. Merely kind of getting to the end of the sentence and understanding it puts you in a in a mode of cain tacitly accepting it to be true and rejecting it as false was a subsequent operation. And one thing you would expect if that, in fact were true, is that behaviorally false statements would be people would be slower to judge false statements of equal complexity than to true statements. And
we found that. So you get people a statement like two plus two equals five, they're slower to say no, that's false than two plus two equals four is true. I mean that that was a it was a fairly significant behavioral difference. But in any case, the.
We let me, let me dive into that for a second. Does that mean that hearing other perspectives, being exposed to other perspectives, which you think, well, that's false, But at least you've spent that moment standing in that person's truth for a moment. So, for example, with political debates, you know, one of the things that I think is the most important for students is to, you know, get a three sixty view of the arguments, to debate from both sides,
and be ready to do that. When we were in school, I think, I don't know if this is retrospective romanticization, but I believe that was more common. You would you would come into a debate ready to do either side, and therefore you knew the issues and the nuance is slightly better than let's say a kid who's in college who thinks, sorry, this is the truth and I'm not
going to entertain another hypothesis. Just based on what you said, it seems to me interesting that if you're told another perspective, you at least stand in that truth for a moment before rejecting it, which might be useful.
Yeah, although it does actually connect with other non normative biases, where it's something called Obviously, we have a replication crisis in much of this psychological science, so I don't know if this work has been replicated. But at the time, there was something called the illusory truth effect, where you merely being told something.
Several times so you're more likely to believe it.
Even in the act of debunking it.
So if I tell you that a certain thing was thought to be true, but in fact it's not true, there's a kind of memory distortion bias where just you know, a year from now, if I ask you about that thing, there's this reliable bias in the direction of, oh, yeah, that was that was true, right, but you sort of the debunking, and this is why misinformation can be so
pernicious and difficult to debunk. And again I bracket this with not knowing whether or not I am now spreading misinformation, because I don't know if this is washed out.
I've certainly heard the misery truth effect many times, so now it's stuck in me as the truth. Ye fine, So okay, So coming back to this then, so Spinoza said, you know, you have to evaluate something as though it's stand in that space for a moment before you make your decision that that's actually false.
Yeah.
So the reporter of that in our study was mostly activity in the anterior insula, which is, as you know, the insula is the cortical structure that that is, you know, very focused on in terror reception. And it's just the kind of the states of the body are being evaluated
here and discussed. I mean, the discussed literature suggests that it's mostly driven by the insula, and it's and so that this was somehow unsurprising to us because it does seem I mean, I think just you can get in touch with this in the first person subjective way, there is something to reject a proposition as false feels a certain way. I mean, it is a kind of emotion, it's a psychological rejection state. And if it's emphatically false, certainly on a topic you care about, there's really just
to put that on. The continuum of disgust was not at all surprising, right, And conversely, true statements were evaluated with the same kind of positively valanced reward structures that we know about in the frontal midline area of the brain.
And this might be surprising to people who haven't thought about neuroscience very much, but generally speaking, we know that all of our acts of higher cognition are are leveraged from areas of you know, areas of the brain which were not designed merely for that for that, you know, not designed by evolution merely to support that higher cognition.
So clearly you have to do math and language and neuroscience and everything else that only humans do using these structures that that you know, even our chimp cousins have.
So when you hear something false, activates some sense of disgust, and use something true, it's beautiful and rewarding.
And there was one other piece that the and this came in a subsequent study when we were looking at political versus and religious versus, just kind of semantic ordinary, more ordinary beliefs. We found that beliefs that there was there was a subsequent study where we which I ran with Jonas Kaplan at USC He's over in Demasio's lab, where we tried to in real time push back against people strongly held beliefs while they were in this skin.
So we'd present them with a proposition which we we knew in advance that that you know, some subjects would strongly believe someone's strongly disbelieve, and we'd try to push their beliefs in the opposite direction, giving them fairly persuasive, but still in many cases kind of just sham evidence.
Right.
So like he's like someone who would be put in the scanner and we would we would ask them to rate on a scale of one to seven their confidence that secondhand smoke is a significant health concern, right, and then we would present I think in this case it was sham evidence against that conviction, seeing if we could
knock get back. And but we did this for beliefs that would be nonetheless strongly held, but around which we wouldn't expect someone's identity to be really anchored versus beliefs that were you know, political or religious, where they you would imagine that would be more resistant to to revising their opinion whatever your evidence, right, And we found that areas in again midline, but now in the posterior regions of the brain, you know, kind of classically kind of
default mode network areas were preferentially upregulated there where and some of these areas are and there's other work showing that they when people are given stimuli that that are interrogating self representation itself, like if you I give you a list of words and ask you which of these words apply it to you? You know, we're like patient or kind or anxious or and then people are if the if the if the task is is this me right?
That's this this the same region that gets activated. So you know, at least in that experiment, we we did produce some evidence that beliefs were people were resistant to modifying their beliefs no matter what the counter evidence was, where the were of a sort that seemed to be invoking kind of the self referential uh processing.
So beliefs are tied to identity.
Not not all but such like like again there's there's there were two buckets. Some could be strongly held. But you know, presumably your identity isn't so tied up in you know, whether secondhand smoke is a health hazard versus uh, you know something about you know, abortion is wrong, or you know, whatever your political conviction was that we analyzed in advance.
So this leads us to the question, I think, if what is belief for from an evolutionary point of view, what are these beliefs useful for doing for for the rest of the brain. One idea about beliefs is that they're trying to build a an accurate model of the world. Another version of beliefs is that it has to do with something social, as in belonging, with identity and the capitlain and all. Study you just pointed two would certainly support that. And then another that people propose that has
to do with emotional regulation. It makes you feel better, more certain, less anxious about particular things, especially I think that would apply to letus say, religious beliefs. And it may not be that there's anything exclusive about these answers, but if we were thinking about waiting them, how do you see the function of beliefs?
Well, I think it's subsumes almost every act of cognition apart from kind of pure kind of sensory engagement with the world. And even in that case, you can you can translate the sensory experience into kind of belief talk. I mean, it doesn't have the we don't it's not mediated linguistically in the same way. But like my belief that that's a table, it's built up over you know,
you can in many channels. One channel as I can see it, and I can see that, you know, supporting your drink, and so we're using that as a table whatever it is. But the linguistic mediation comes into play when we might say, oh, would you just would you go put that over on the table. It's like then, so now I have this abstract representation of the table, I walk into the room, I find the table, right.
So obviously we're engaging this in multiple modes. But once you break free of just moment to moment, you know, behavioral imperatives of navigating that, you know, our mediate environment, it's almost all linguistic or you know, imagistic representations of possible states of the future, you know, plausible states of the past. You know, do we try to reconcile our memory, you know, our memories when there's any kind of discordance there. But again, we're just trading in language. Oh no, no,
he didn't say that. He said this, Or no, no, that wasn't. That wasn't on Tuesday. That was on Wednesday, And I mean it was It's all we're doing is talking to ourselves and talking to others about what the world is like and what our place in it is and what should we do next. I mean, we have just we have this massive navigation problem which covers really
our life in every respect. I mean even you know, I view morality as a navigation problem, and morality really is a story of what should we do next, you know, or in the future, right, Like, what is what's the right thing to do? So you get two people in a room and they're for whatever reason, their incentives aren't aligned, and there's some kind of negotiation, and then it immediately becomes a conversation about what's the right thing to do, what's the fair thing to do, what's the what do
we do last time? And if you actually kind of drill down on what's happening at each point along the way, and what is anchoring everyone's emotional response. You're looking at knowledge claims about the way the world is, the way that it's likely to be, and people's kind of the waitings, the emotional waitings. People are giving these claims with respect to kind of a probabilistic sense of the likelihood that they're true or false. Right, So, if someone says to you, oh, no, no, no,
that's not what happened. There's a very controversial shooting in the news right now that has kind of shattered the country in the last twelve hours. An ice officer shot a woman protester in her car. You know, millions upon millions have seen the video and have rival interpretations of the video based on what they believe is plausible to think, based on the angle they were looking at and etc. All of this is a story of rival beliefs framed
by different shadings of facts. But again, facts in this case are beliefs upon beliefs upon beliefs that people are granting more or less credence.
Right.
So, Like in this case, you know, the woman was driving her car toward the officer, So was she trying to get away or was she trying to run the officer overw how you interpret that maneuver changes your sense of the ethics of it completely. What is an officer supposed to do when he's no longer physically in danger with his gun, but his gun has drawn, he's already
shot one round at the suspect. You either know nothing about that or you know a lot about that, and you have very strong convictions about what the right thing to do is or what the legal thing to do is. Again, is just a blizzard of propositional claims, linguistically mediated propositional claims about the way the world is or should be, And it's all a matter of belief and our creedence around.
It, agreed. But what I'm interested in is not so much the individual beliefs that we might have about the trajectory of the car so on, but the social aspect of this, in terms of what do my friends and my colleagues on whatever my political side of the argument is, what do they believe is true? We're so obviously influenced by our groups, by our cultures, by our families, by our neighborhoods. To what degree is belief fundamentally a social issue in our social species?
Well, I think it is almost entirely. When you look at it's kind of the primacy of language in shaping these beliefs. So we learn the language is a social phenomenon, right, It's an intersubjective phenomenon. You learn it. I mean, it's not that we don't have a propensity to learn it. Obviously we do, and that is kind of a private
individual fact about each of us. But when you look at how it gets trained up in dialogue with your parents and caregivers and other people in your environment from your first hours of life, it's an intrinsically social phenomenon which we then later internalize, such that even when mommy and Daddy leave the room, we're left talking to ourselves. Right, And so there's that kind of inner voice that gets
kindled and which we never seem to shake. Most people are mediating all of their experience with language, and we do this with one another. It is the basis of virtually all human cooperation. I mean, it's certainly cooperation with respect any complex task. I mean, like if you got up and started a trip, you know, I might wordlessly and instantly you know, grab you to keep you from falling over, or you drop something, and I try to
catch it. That's not linguistic. But basically, any cooperation more complex than that is a matter of us talking to one another about what our goals are and what we're going to do next.
Yeah, So how do you think that applies to beliefs? Though? My belief about whether the woman was trying to kill the cop or get away, my belief about gun control, abortion, whatever, What do these things have to do with the people I'm surrounded by? In other words, how do we make our beliefs about what it's true what it's not true.
So all of our talk about beliefs is occurring in a community, right, in the community of people who are we have direct relationships with, and then just a community of that is just a matter of are being embedded in an information landscape where we're hearing from other people who we may never meet, or we're reading books, you know, written by people who may have been dead for hundreds of years, but we're trafficking in the kind of just
a torrents of linguistic representations of the world, and we're constantly revising or resisting revising our view of the world
on that basis. Right, So you read something and you know this is where various you know, reasoning heuristics, you know, normative and otherwise come into play where you could have something like confirmation bias, right, which you know, in science we're alert to the fact that this is a kind of you know, cognitive defect or failure mode where it's like, you know, if you're just looking for the confirmation of your cherished hypothesis or your cherished opinion, well then we
know that's just not a good operation to see if in fact you're you're in touch with reality. And so, but in every area of discourse, whether it's a formal one like science or it's just we're just you know, gossiping with friends, we're constantly trading in statements about the way the world is. And these statements could fall into various categories that are just are more or less morally salient or or psychologically.
Combustible.
I mean, so, you know, if you if you're talking about religious beliefs, like core religious beliefs that are you know, foundational for a specific community, obviously those have a different gravity to them, and they are all you know, pathologically in my view, they are protected by certain norms that are really the antithesis of what we what we do in science where you know, you're I mean this, you know, religion is the only area of culture where the notion
of dogma is not not pejorative. R I mean, dogma is literally not a bad word in you know, dogma's you know, traditionally Catholic, you know, and word. These are the things you're gonna believe without you know, sufficient evidence, and you're gonna believe them as much as you can believe anything, and they're not on the table to be
argued against or revised right now in science. If you if you can say, if you tell someone well that's just you know, you're being dogmatic, or that's that's a dogma, well that's not you know, that's definitely not a nice thing to say about somebody's reasoning. So yeah, so this
is this is it's it's everywhere. And again, the hypothesis going into this original uh neuroscientific work was that there's something importantly similar across across all of these domains that we're doing where you're you're just you're you're accepting a representation as behaviorally valid or you're rejecting it as not and or you're or you're in a kind of holding pattern where you just can't know one way or the other.
But I still want to understand what is the difference, either from neuroimaging studies or just how you've thought about serveral years, What is the difference between let's say, scientific and political and religious beliefs. That are all beliefs that you might take to be true or false. But it does feel like the social component is a really important
one to political beliefs, maybe scientific beliefs too. In some cases, the emotional regulation might be really important to religious beliefs, as in, I've got a lot of anxiety about this, and this makes me feel better if I were to accept that. So there are these different components about why we believe. In other words, if we take the position that belief isn't actually just about figuring out what's true in the world, which we wish it were, but it's
clearly not that in brains. So this is my question for you. How do you think about the when beliefs are for the purpose of social cohesion, when they're for the purpose of emotional regulation, and when they're for the purpose of getting the right answer for predictive actorscy the world.
Well, I think all of that is still pretty knit together. But yeah, I would I would agree that certain beliefs function, or certain professions of belief at least functions kind of you know, loyalty tests or signaling as a you know, you know, a shibboleth as you know that where you're signaling group membership, right like like we have a code, and on that basis, there's a level of kind of social trust and expectations about each other's ethical intuitions and
political commitments. And it's the kind of shorthand. But if you're going to tear at those sort of core group identity assertions, whether it's a you know it's a religious you know, a mainstream religion or you know, small or still like a spiritual or religious cult, or a political cult say, or political culture, these are kind of anchoring assertions about what is true about what really happened, about what is real about the unique sanctity of a given book, say,
which we're not disposed to rein.
Tear very much.
And when you when you talk about specific things where they really were, these different ways of knowing collide. There's so many examples. He takes something like, you know, the shroud of Turin which was thought to be this. Perhaps in certain circles it still is thought to be that this Christian relic where this shroud had had received the facial imprint of Jesus, you know, it was thought to be his burial shroud. But then carbon dating was done on it and it was it was revealed to I think,
I think it's the thirteenth century relic. So it's you know, it's kind of a thousand plus years off from being the real shroud. So is this a claim about about purely religion or is this now? I mean, once carbon dating gets invented and you can you can use it, it suddenly becomes a claim about about chemistry, right, you know. And so I mean that is the the enduring tension
between science and religion. Whenever you're making claims about the world, about the past, asked about what happens after death, that the way the mind works, about the connections between people, you know, tangible and invisible, you're at least in principle, in principle risking making claims that wander onto the territory of every other way we judge the truth value of claims about the world.
So why is it so hard for let's say highly intelligent people to come to agreement. I mean, we know in any science department in a university, people argue. They spend their careers arguing. Again, if this were just about the brain trying to find predictive accuracy, that would be very difficult to understand. But because of all these other components to it, the social, the emotional, it's it's a much more complicated thing.
The search for truth it is. I mean, you can sharpen it up more than we tend to. For instance, you can you can be alert to the problem of someone making claims that are on falsifiable. Right. This is something that that Carl Popper, the philosopher of science, gave us just the the litmus test for him was the falsifiability or lack there of a statement that was That was how that was the boundary between science and and the rest of our thinking. So what doesn't mean to
make a claim that is unfalsifiable? It means it means that you're you're saying something is true. And yet there's no state of the world that could know, even not even in principle, a state of the world that we can imagine that would report to the falsity of the claim.
Well, then you're it is.
What's an example, you know, this is a the world as the entire world, the cosmos, everything that is apparently real was created by God just you know, ten seconds ago. But what God also created were all of the memories and all of the backward looking evidence we think we have about the antiquity of the cosmos. But the cosmos is actually ten seconds old. We just have all these memories jammed in our heads and all of this stuff
on the on the internet, et cetera. And the rocks or the way they are and that, you know, so nothing would be different. But you know, this is this is my claim. This is my magical claim, right, so it's unfalsifiable. You know, every time you're going to give me something that you think would falsify falsified, I can
just kind of layer into that into the thesis. Well, no, God made it that way so as to make it seem like, you know that way, right, So we're not We tend to not be interested in claims like that for good reason, right, because they're just they're not one they you know, there's they run against the grain of
a very useful heuristic. We have in science, which goes by the name of Oakham's razor, but more generically, just the idea of that a thesis should be in so far as it's possible, parsimonious, right, Like you shouldn't just proliferate your you know, claims, you know, so as to shore up I mean, we don't want epicycles within epicycles within epicycles. We want the most elegant, you know, simple thesis that still conserves the data.
But I still want to understand. If you got let's say, Popper and Spinoza in a room together. Here, you've got two very clear thinkers, but there would be beliefs that each of them hold that they would end up disagreeing on. And again I'm just I'm just circling back to this point that it's it can't be just about the rational. There are all these other reasons why humans hold beliefs, you know, the social and emotional or the things I
keep hanging on about. But this feels important to me in terms of understanding why it is so difficult, why there's such polarization. Because what would be great is if we could sit down and you know, make our statements shout everything in all caps on X or blue sky, and everyone sees the logic and says, oh, okay, we agree, and and the whole world would come to agreement, but
that that will never happen. And and I've just been curious about why why beliefs are so different in different heads and often so intractable.
Yeah, well, people have I mean, obviously the people are driven by more than just a desire to not be wrong right, or to have or to to have their their representations of the world be accurate in the abstract. I mean, people are incentivized by the things. You know. It's like, you know, there's the the famous line that now I've actually forgotten how it goes, but something you know, it's hard to convince a man of of something that one of his occupation requires that he not be convinced
of it, or something like that, right, you know. So it's like it's like, yes, if you're if you stand to lose a lot of money if a certain a certain state of the world is so, you're going to tend to avert your eyes from that for as long as possible, because you're I mean, this is just the the classic problem of and says I mean just it's a perverse incentive.
Right.
So in science we have worked this out to a remarkable degree. I mean, the thing that distinguishes the culture of science. It's not perfect, but what distinguishes science from basically everything else human beings do is that we have created a culture where there's a competitive apparatus that is error correcting by its very nature, right, Like where you know, we could both be in neuroscience and have rival theses and we're trying to you know, could falsify each other's
claims or at least pressure tests them. And science is the only part of culture where you can actually win points for proving yourself wrong, right, I mean, just, I mean it doesn't happen as much as we would want
it to, but it is. It really is a kind of a gold standard moment of scientific communication for somebody who has had a cherished thesis for you know, decades, to admit he or she was wrong about basically everything in public and to thank the person who did the work that showed that he or.
She was wrong.
Right, I mean that you know, people people have you know, again, it probably doesn't happen in real time as much as
we would want, but it certainly happens. And it is it really is the quintessence of the scientific enterprise, right and and and what what is not the quintessence is to see some scientist who is just clearly emotionally attached to all of the sunk costs of his work, you know, his work and his reputation, and he's been known for this thing and this now, this battered thesis, is losing
his credibility among the next generation of scientists. And this yet, this old warhorse, is holding on to this doomed project despite the fact that there's a now mountain of good argument and good evidence piling up against him. At that moment, it's actually right, you know, it begins to seem more correct to say that what he's doing is it's not
like an alternative version of science. It's like it's like no longer science, right, Like you're you're failing these these obvious tests of credibility and the obstacle course has become too complicated for you emotionally for some reason, and you're no longer navigating it.
Yeah.
So okay, So this brings up something about science that I flagged earlier that I wanted to come back to which is so when I read these original your papers from two thousand and eight and two thousand and nine. Right, So there's there's a belief that you think is true, belief that is false, and then the uncertain ones that you just don't know about. But what's so interesting to me? As we know in science there's instead a waiting or a probability that we put on different sorts of beliefs,
and even our most cherished beliefs. We never use the word truth in science. We say, Okay, the way to the evidence supports this, and if in a few years the way that the evidence might support something else, fine we go with that. And so I'm curious how you would I know this wasn't part of the study, but how you might think about beliefs of this sort, which is the type that we hold in science of saying okay, well,
there's mountain of evidence here. Some of it points the other way, someone points off this way, but most of it is pointing over here.
Yeah.
Well, I do think that virtually all of our beliefs are probabilistic. I mean, certainly in science explicitly so, I mean we have been we have been beaten into shape on that on that point, then again by largely by Popper but by others where we just know we know enough now to not say that's you know, we're one hundred percent certain something, and so it's just the way of the evidence, et cetera. Yeah, I mean, just in
very simple terms. And you could imagine being absolutely certain, based on your or understanding of mathematics, that every triangle has one hundred and eighty degrees.
Uh.
And I'd say, well, you know you're you're certain certain and yes, absolutely certain. Would you be willing to bet your life that it's so? And you might balk a little bit and say, yeah, yeah, i'd be. If i'd better on anything, I'd better on that. Well, it's it's sort of like two plus to equals four for me.
But then someone introduces to you a concept of you know, non look Euclidean geometry that you didn't know before, right, So like you've just what you had in mind was a triangle on a flat surface.
But now the basketball.
Yeah, but now someone like Remond comes along and says, yeah, just visualize this. We're now going to curve the space. What's happening to that triangle on the basketball doesn't look like a little bit, a little more, a little less one
hundred degrees. All of a sudden, your intuitions get knocked around, And if you're rational, you are now going to update your file about triangles, right, and so if if a core belief about something that's simple as a triangle can be overturned, But you just had a further utterance from you know, somebody who knows more about math and you, what is truly off the table for revision is probably not much.
I wish that were true.
I think I may be more pessimistic than you on this point, which is, you might easily convince me about a triangle, But core beliefs that I might have on any hot button political issue, those are tougher. I'm using myself.
Can give me an example, Oh, whatever.
Your views are on abortion or gun control or whatever, those are resistant to change because it's not as simple as a logical thing of Oh, I see the triangle on the basketball. Got it, I've now changed my model.
But I'm sure certainly in both of those cases there is much more information that could be forthcoming on either topic that could push someone's because because people's intuitions, especially on those two topics tend to be anchored to like some meme that got in their head, you know, very likely early in life, that doesn't actually have a lot of content to it. So it's like the claim that abortion is wrong. That that that, you know, I'll drill
down on that a little bit. It felt it's likely backed up by utterances like life starts at the moment of conception.
Right. That's like if you.
If you're if you're a you know, a religious believer who believes abortion is wrong, Well, that's it's just that's not true. If you're if you're a Muslim, that's just not true. You know, in the Muslim faith, I believe the soul is thought enter the zygote. I think around day eighty. It's either day eighty, day one, twenty something like that. So but definitely not at the moment of conception.
Right.
So, Okay, perhaps you're a Muslim who just didn't know that that was the doctrine, right, so someone told you life was started the conception. But then but then they actually check that the Holy Books and and and it turns out you're wrong. Okay, that's kind of like triangles all of a sudden. So it is with gun control. I mean, people just don't know a lot about guns.
They don't know a lot about the actual statistics of human violence, and they don't know how many people get shot and how they get shot and why, and they just they tend to have they just I mean, it's just there's just a it's a thicket of ignorance and assumption on on lots of these topics.
And and I'm not.
Sure I agree with this in the following sense, just because I've I've known very closely family members and friends on very different sides of this issue who are extraordinarily knowledgeable and bring up an infinite encyclopedia of statistics about these issues. But they're on opposite sides, right, So so ignorance, isn't I think a way that we can explain away some people's positions on this.
Well, I'll grant you that it is possible for two people to have more or less the same set of facts and to feel very differently about them. And then the question is what explains that?
Right? So what else? What else do they believe? Right?
What's about their internal model in this world and their expectations about the future and the government of the future, and the crime of the future and so on that makes them feel that, hey, I don't need a gun to a house. I do need to gun to a house whatever it is.
Yeah, So in that case, I would say they're waiting different probabilities. I mean, so, it's almost never the case,
especially for that debate. It's almost never the case that you're talking about two people who are in touch with the same set of facts, because it's just, you know, the people who are deep into gun culture, who are it's all about the Second Amendment, and they're happy, they love shooting guns, they have guns, they're comfortable around guns, they know they can store them safely, say, and they have a very They've spent ten thousand hours rehearsing scenarios
where they would have to defend their lives or the lives of their loved ones against some malicious intruder. And that's there, you know, that's become, you know, a deep part of their identity. Even that person is dealing with
a very different set of facts. And the person who has no experience with guns, doesn't want any experience with guns, hates guns, thinks they're dangerous, thinks that it's just appalling that we live in a society that has so many guns and we've got this absurd level of gun violence that's unrecorded, that that makes us an outlier of an outlier in you know, other in the developed world than just to kind of it just just we're just as a horror show.
How did we get here?
It's it's like just pure masochism that we were living with this, you know, the religion of the Second Amendment. Right, So those two the people, no matter how well informed each representative of those two camps thinks they are, I guarantee you that in almost every case, they have very different beliefs about what is actually true in the world with respect.
To the races of violence.
It might both be holding on to facts, though they're paying attention to different facts. So the gun owner is paying attention to things like tyrannies of governments and what happens when governments take guns away from the citizens and so on. Those are the kind of facts they draw up. The gun non owner is drawing up facts about crime and how many accidents happen in households with guns and so on.
But they might have those facts wrong. So like the non gun owner will believe that it's almost never the case that someone successfully defends himself with a gun or defends his family with a gun. That just doesn't happen. That's just NRA propaganda. What really happens is kids accidentally get a hold of their parents' guns and get shot and shoot themselves or shoot someone else, and it's a tragedy, right.
And the statistics.
Around gun violence are highly politicized, and it's very hard in fact to go in with an open mind and figure out what is actually just real, you know, empirically, because it's been so polarized. But even if we create the perfect case where they really do have the same sets of facts, uh with respect to what is being talked about, there are other facts and other propositional attitudes beliefs that are covertly working to change the sort of the waitings that they're giving to all of these facts.
So like there are people who are just for whom the prospect of being finding themselves helpless to defend themselves
or their their loved ones from evil. That is the that is an outcome that is so bad, I mean, the utility function that the that is that is so much worse than all the other bad things that could happen in life that whatever low probability you put on it, like the probability Let's say we could agree about the probability of a home home invasion, you know, where the person has come to murder you and your family, Right,
whatever that probability is. If I wait that as worse than everyone dying in a fire, and and and a hundred other things that you think are probably worse than than than a home invasion, then we just have we just have different utility functions. Yeah, right, And that's and that is often at play. I mean, that's where emotion
comes into it. Where you see people get really you know, we seem like we're just having a conversation about facts, but you see people, even granting the same sets of facts, become highly polarized and energized around the significance of those facts.
Yes, yeah, I couldn't agree more. And this is why I feel like I'm still trying to figure out how to think about beliefs because so much of our positioning
on everything has to do with these other issues. And I've I mean, I watched this all through the last decade where people will end up switching political allegiances not because they like the fact, so much of the other side it's because something discuss disgusts them about their own side, something gets said to them, or someone accuses them of something and they get so mad about it, and that's what drives them.
Well.
But politics, I mean, one thing that's so frustrating about politics is that it should be the space in which we we rationally operate so as to cooperatively produce outcomes that that you know, improve our lives.
Right. So it's like we politics really is.
An art of cooperation and yet and to do that, well, we really have to be mean, you know, reason has to be the master value, and what's true, what's plausible?
Uh?
And then the ethics. You know, you can't keep ethics out of it. I mean, there are questions of what's fair and what's et cetera, what is what is good? What should we value? But so much of it is pure tribalism, right, And it's pure in group signaling and outgroup disparagement, and it's and and there when that is the becoming the master value, it's just it's just not
about honestly talking about the world at all. Right, You just like, like, I mean, in so far as you're trading and claims about you know, the purport to be fact based, it's still what you're really doing is it's a team sport and you're trying to figure out how to win, Right, You're just trying to So if you're saying negative things about the opposition, you begin to not care whether they're even true. Let's just see what can stick, Right, this guy's a you know, this guy's a child molester. Well,
what's your evidence for that? I you know, I don't know. I saw something somewhere on social media. But you know, like if I keep repeating it, you know, enough people are going to think he's a child molester that it could harmless chances come election time.
Right.
So it's like that's the game, and it's awful. Right, It's just so degrading in principle because it's leveraging the most anti social emotions. Mostly it's leveraging fear and outrage. I mean, I get irrational fear and outrage and false certainty or the pretense of certainty where you actually have no rational basis for certainty.
Yeah, And in fact, this reminds me I want to come back to something at the very beginning about a belief that one might think is false. You found that that activates the anterior insula, presumably a sense of disgust about that. I wonder the degree to which this might
differ based on different false information, some false information. For example, you see something online clearly misinformation, disinformation, and one might think this is really a concern for my side of the political argument, or for my you know, ethnic group, or for my anything this kind of misinformation. I'm simulating the future. What if everyone believed this piece of misinformation
that's really disastrous? And so I wonder if some sorts of misinform, let's say, false beliefs, our reaction to them isn't simply discussed, isn't simply ah, okay, I've categorized that as false, but instead it's hey, I've categorized this as something that's really a concern for me.
Yeah.
Well, it's very easy to say. Again, this comes back to the power of ideas, when you're thinking of just how what is what could the effect be of a single sentence? You know, in any language, the sky really is the limit. If the front page of the New York Times reads that nuclear war has started in thirty minutes,
it's going to get to you. Right, that's everything, right, This is like, I mean, we're just decoding words in this case on a website, and yet unless we can figure out some reason not to grant them credence, right, it's like, okay, wait a minute, this has been hacked, is it? You know, is it April Fool's Day? There's something wrong with my computers? Like this is really the
New York Times? Is that really the website? Now with AI, were on the cusp of having to declare, you know, epistemological bankruptcy with respect on the internet because any video can now you know, virtually be faked. Right. So even now if we saw, you know, a video of Vladimir Putin saying I've just launched you know, our full arsenal the evil Empire of the United States. It'll be there
in twenty four minutes. If we saw that and it was translated perfect Russian, and you saw you know, Gary Kasparov saying yeah, that's that's really Putin and that's really Russian and that's really what he said, and it's now on the New York Times website. Now you're thinking, wait a minute, that does the New York Times have sufficient technology to detect deep fakes? Do I really have to
believe this? But once you can't grab a handhold to resist this slide into just helplessly imbibing the propositional import of those words. Right World War three, you're emotionally and behaviorally completely exposed to the implication of that claim.
Oh that's interesting.
You're exposed to it, or you've put up walls against it as well.
The way to put up walls against it is to find I mean that there are other If it's big enough, you know, it becomes irresistible. But I mean you can there are many ways to put up walls against beliefs we don't like the taste of, right like, we can ignore them, we can move on, we can distract ourselves. You know, this is this is kind of the purview of what we think about is self deception, right Like, how is it? How did how did the guy not
know his wife was cheating on him? I mean there were these signs and he sort of I mean he must have seen that thing or heard that thing, or if she did that and I saw that. So there's all these you can you can say that there's this sort of this this boundary even within within the precincts of one mind, where you can we might say there's a an avoidance of knowledge when the knowledge is really there. But no if so if if the whole world kind of sticks your face in a claim. Right again, it's
on the internet. The Internet's on fire with this thing being news, and you have to figure out whether it's true. Again, it's really just a matter of words or images that
purport to represent the world suddenly transforming your life. Right, and this is all of a sudden, This is like like literally a sentence could be spoken to both of us that was so incontrovertible given the framing around it, given the way we both know the world is given how much would have to have been faked in order for it to come that way to us from that source, right, and this goes to the role of authority and gatekeeping
and right. So, but provided it comes you know, over the transom at just the right angle, you know, the distance between us in this moment, you're reasonably comfortable and without much of a care in the world, and absolute panic is just you know, it's not much longer than it took to decode the sentence in the first place. It's like, holy shit, right, I mean, it's rare that that the stakes are that high and the testimony is that, you know, unimpeachable and unequivocal and irresistible.
Right.
So it's where we're more often trading in the in the land of well, it doesn't matter that much whether it's true or false. You know, I'll take it on or not. I can act as if it's true, but if it turns out not to be true, and so we're and and and we're we're dealing in probabilities where yeah, we're willing to bet something in that direction, but we're not willing to bet everything in that direction.
What I mean is when you see a piece of obvious, blatant misinformation online, someone posts something stupid and and it feels like what your mind does is simulate, Wait, if everyone thought this, if this spreads, if this meme goes viral, that's going to have consequences. Yeah, That's what I'm pointing to. Is not necessarily like the nuclear war thing. But here's
an example. Right after the Bondie Beach massacre of Jewish celebrating Hanukkah, somebody posted something saying, oh, it turns out one of the shooters is Jewish, and they showed a Facebook page of the guy wearing a yamica and all his friends and his friends on there. It was such an obvious AI fake when you zoom in on the buttons and so on, the buttons didn't make sense and so on. But I think the feeling that one might have viewing that is, hey, that's a dangerous piece of misinformation.
Why because if it spread, if lots of people saw it, blah blah blah. So what I mean is it's not just the evaluation of, oh is this true or false, it's the I'm just wondering if the discussed element in real life might be the what are the future consequences of that piece of mistass?
Oh yes, well yeah, I think that can That can make the field of disgusted not only a neurophysiological analogy. It really can make it quite salient. I mean, if someone says something that you know to be false or you have every reason to believe it's false, and not only and it's and and if but if believed, it will open the door to just manifest harms in the world.
And it's just it's it's an ugly divisive idea that you believe was maliciously concocted so as to produce those harms, right, So like it's not all these all this negative uh, negativity stacked on top of it. Yeah, I mean, I you know, I feel discussed all the time, and consciously feel discussed all the time at what people pretend to believe, you know, or what some people actually wind up being taken in by and actually believe. And the consequences in
the world are obvious. I mean, I mean, in fact, if you once you put this kind of lens on to your your view of more or less everything, Uh, people, do you just see that all of the world's mayhem is a story of kind of failures, I mean, apart from the things that are visited upon us from nature, just like bad luck, you know, just like you know, an asteroid impact or or a virus you know, jump
species and now you know we're suffering its consequences. Everything else is a self inflicted wound based on bad ideas, you know, just just rank tribalism and conspiracy thinking and lies, I mean conscious lies meant to divide people and to motivate them to produce harm. And it's it's all a matter of kind of this belief making and belief grasping machinery of our minds that is just ceaselessly doing its work.
It makes me wonder why changing our minds is so difficult when it comes to a scientific proposition. It seems to be easier, maybe because often those are like chess moves in the sense that they're cognitive. There's not a lot of emotional weight with it. But like the example you gave of the older professor who's emotionally so weighted down by what he's come up with, sometimes it's very
hard to change our minds about things. What's your take about the challenges and any hopes about people changing their minds on things.
Well, I think you need one needs to internaliz. I mean we do this in science, but I think this just should be exported to the rest of culture. I think we all have an ethical obligation to internalize a value that is deeper than any place where our kind of wishful thinking or the way that our preference is for the way the world is could be anchored right, And the deeper value is we actually want to know what's real. Right, We would like like we have a
reality bias. And it's not to say that that you can't find these sort of low instances where a person's interest might be better served by not really being in touch with reality, like some some scope for delusion, you know, self serving delusion might you know, we can certainly concoct those examples where like the person would have if he if he knew what the people in the room really thought of him, well then he wouldn't perform nearly as well as he did when he thought he was popular
or you know, thought he was handsome, or what everything was. But generally speaking, we shouldn't want to be deceived or self deceived or psychotic or and and trading in these hallucinations and having and watching our culture get bent every whish way by these kind of mass adoptions of delusion. You know, there's moral panics and paranoia and lives and half truths that become operative.
You know.
Again, politics is so much the story of this. But it's you know, I would say it it embraces you know, almost everything we do. He's just I mean, it's not it's not that understand in the world is everything. It's not that being rational is everything, but it is this, it is the thing that can safeguard everything else that
we care about. R I mean, I view your reason as, among other things, the guardian of love, right like like like there's there's other things you can love, like the other experiences you can have that are not a matter of just rigorously understanding what is so right, but rigorously understanding what is so is the thing that get that that that that stands at the gate to protect all those other moments where you just want to you know, hug your lover or you know you're your best friend,
or play frisbee, or appreciate art or just watch a movie. Well, when it comes time to figure out what's actually happening in the world, when there's a pandemic raging, or when you know people are flying airplanes into our buildings or or whatever the whatever it is, we should have a very deep bias for the truth.
How would you teach that to high schoolers? What would you tell them?
Well, one, you want to remove the stigma around honestly confessing that you don't know what's true, right, So saying I don't know should they should have absolutely no discomfort attached to it. In fact, failing to say that you
don't know when you clearly don't know. Should should That's where the stigma should be, right, So to pretend to know something you don't know is where I mean that that should be you know, scored as kind of unseemly and then people should get the social feedback around that.
So whenever you were caught bullshitting, right, that's the thing that should be embarrassing, not when the teacher asked you the question and you just didn't know, right, or And people should never be afraid to ask what seems like
a stupid question, like even in a scientific context. I mean, you know, there are great accounts of, you know, the behavior of some of the most eminent scientists you can think of where one of the things that made their company so EDIFYE and it actually kind of discombobulating for you know, lesser scientists is that they would ask a lot of dumb questions, you know, apparently dumb questions which if if if a novice was that would ask that question,
they might actually see there might be some kind of social embarrassment around around that's kind of a dumb that's really an elementary question. But when you know, Richard Fyneman is asking you that dumb question. You know, you know you're in the company of Richard Feynman, right, you know he knows physics at least as well as you know physics.
And he's asking that question. He's actually now putting pressure on what is kind of a load bearing wall and may in fact be a load bearing fiction of the whole enterprise. Right, And so you should be not at all afraid to ask dumb questions. And there should be this kind of master value where you want to calibrate your conviction rather finally with the weight of the evidence
and the and the and the solidity of the argument. Right, like like you you want to you don't want to be at eleven with your certainty when there's actually nothing to stand on. And you also should not want to be wrong for a moment longer than you.
Have to be right.
So like so like if you I mean this is this is the spirit in which I would have any debate on any topic. I mean, I have you know, I tend to know what I think when I'm certainly on a topic with where I'm in any kind of
debate mode with somebody. But if someone makes a point where I see their right and I was wrong, like my my kind of master my master value in that moment is not to defend myself, you know, just to put a brave face on my on my error and try to, you know, just kind of beat back the truth that they just articulated, because I mean, there's for two reasons for this. One is if the place I'm standing actually has no support and the person who's criticizing me can see it, and they pointed out, and the
audience is likely to see it. If they can't see it in the first second, they're likely to see it like thirty seconds from now. What I really want to do is no longer to be standing there, both as a matter of like, you know, like just intellectual integrity, but also just as a matter of like not losing you know, egregiously, if I in so far as I care about the way I'm being perceived by the audience. So the master move there would be to actually let
the person educate you on that point right now. Hopefully it's not a point that is so central to to you know, that the topic you were debating such that you now have just admit you're completely wrong and you've changed your mind. But in fact that were the case, that's what I would want to do. I mean that again, that's the thing. I go back to the scientist who might have spent his entire career banging away on some doomed theory. The moment has really proven wrong, right the moment.
I mean again, this is these are situations that are few and far between, but it does happen. The pivot you want to be capable of making is just to be grateful to now know what's real, right like that. I mean, that's where the reality bought. But we should be hungry to wake up from the dream of misapprehension. This is a hallucination that this is a spell we have to
keep breaking. We just we come into this world not knowing what is going on, and our entanglement with it and with others and with culture and with increasingly sophisticated ways of knowing is is continually trimming down this, this misunderstanding and aligning us with something Again we're not We never get some final account where we can step outside of our view of reality so as to look back at it and say, oh, yeah, this is where you This is exactly how the map fit the territory. We're
always just kind of living the map. But when we when we you know, bump into a hard object, you know, in the darkness, we should know, we don't want to do that again. But it is, it comes down to a very clear sense of what intellectual integrity is. And we don't teach that very well. We teach it well in science and in you know, kind of rigorous areas of academia, like you know, philosophy, and in other areas
we seem to teach the antithesis. I mean, I mean, one of the worst things that's happened of late, and it's it has been a source of great cultural division, is that so much of what purports to be knowledge gathering in the Ivory Tower has become just rankly politicized, such that you have whole disciplines which are have just been vitiated by politics.
Right. So this it's a culture, a culture.
Of activism that is masquerading as a culture of of knowledge acquisition, and it's completely upside down. And you know, therefore, and it's been it's you know, the the revolt against the elites is all too understandable in light of this, and it's I mean, we're going to take a generation to recover I think from.
It, I agreed.
I will say, by the way, it is a total side note, I'm very hopeful about the role of AI helping with this problem. Here's why we can set up a debate bot so that each student debates the bought some hot button issue. They go, they get graded on the quality of their arguments, and then they switch sides and they argue the other side. Why because the AI
is great at doing this. It's calm, it's patient, It's able to do that with every student for as long as it wants, whereas no teacher could ever have that kind of time or patience. So I think this is going to be really helpful in terms of teaching students this three sixty view. Is there anything you've changed your belief about in the last decade or two?
This is previously actually this part of the AI conversation. So I've been very worried about AI and the fundamental
question of the alignment problem. I just what it's going to look like to build you know, truly superhuman general intelligence and kind of the ways in which we might do that wrong, you know, and effectively you know, ruin everything, because suddenly we have we have created a technology that is more powerful than we are, an autonomous and not aligned with our well being, and we're kind of negotiating with it now, and it becomes this doomed chess game,
Like it's like, you're not going to play chess harder so as to to realign the chess engine. Then now is now just beating you at chess, and you're not going to negotiate with a super intelligent AI that's not disposed to take your advice hereafter. So, so I'm worried about AI. And I did think at least five ten years ago that autonomous weapons would be something that we wouldn't want to do. That just seems like a very dangerous thing, And so I think I even signed an
open letter that we shouldn't have autonomous weapons. I think, my again, this is not really a happy epiphany, But I do think that leaving the alignment concern aside, given the arms race with China clearly and other potential adversaries around AI, I feel like we just need to win that arms race. So i'm, you know, until I hear a better argument. Now, I'm in favor of all the ways in which AI can be adopted by the military.
Autonomous weapons included. I just think the failure mode of declining to do that work and waiting for China to do it is so obvious and excruciating that I'm willing to roll the dice on let's get there first. The first time I heard about it, I was sort of like the gun control guy, like I don't want anything to do with guns, and they're just dangerous and awful. And then I sort of just flipped to none, no, no, we need as many guns as possible, and we need
them fast. But again, I've obviously opened to argument on that, but that's that was a very clear change of heart.
This comes back to the central theme that we've been discussing about how beliefs you know, in the laboratory, the way one has to study it is with simple beliefs. In the year, they're true or they're false, But in real life they're tied to everything else, including our views of larger and larger geopolitical issues, such that you might have a flip of a belief predicated on these other
larger issues about where you think the world is going politically. So, given this other constellation of beliefs that people have that navigate their particular beliefs about politics or abortion or gun control or whatever. Is healthy disagreement something that you see as possible, and if so, in what ways would you like to see that happen?
Well, I think we we live in the zone of healthy disagreement all the time, and we're constantly disagreeing with with people we love about things.
The question is how much.
Does it matter? And when it matters, it's like when you raise the stakes, then you begin to get uncomfortable and you feel like you have to converge. I mean, it just became it becomes behaviorally and emotionally imperative to converge, certainly when lives depend on it. Well, then all of a sudden, then it's like it's a kind of a moral emergency to get get your facts straight, get you know, you know, you get some of the best argument and win.
So it really just it comes down to what the stakes are and whether we're in a zone that's anything like that which is indicated by the common phrase, or we're just gonna have to agree to disagree, right, Like, well, yeah, that's we over here. We can we can do that and still be friends and this is perfect, right, there's no problem like that's kind of the spice of life to agree to disagree about that. But over here like we can't take another step until we figure out who's right.
That was my interview with Sam Harris. Sam's research here shows that even at the most basic level, the brain distinguishes between accepting or rejecting a claim or remaining uncertain about that claim, and that distinction the brain makes that then roots what you do in terms of feeling and behavior. Some things you act on right away because you believe they're true, and other things you dismiss and they don't
move any further through your neural system. So your brain's assessment about the truth or falsity of something determines everything about what happens next. As Sam also mentioned, beliefs differ in terms of how tightly they bind to identity. Some beliefs are pretty easy to revise when new evidence appears, like here's data about how you could make nuclear react safe. But other beliefs are anchored to social belonging or moral commitments you have, or longstanding narratives about who you are
and how the world operates. In those cases, changing a belief can feel like changing sides or stepping away from a community or giving something up that wants provided stability. What emerges is a picture of belief as a kind of navigation system. It helps us decide what to act on and who to trust, and what to fear, and what futures to prepare for. And belief is shaped by evidence and facts, yes, but also by incentives and emotions
and social environments we inhabit. Zooming out this brain research has consequences far beyond individual brains, because the same mechanisms that allow us to cooperate and reason and build shared models of the world can and also amplify division and harden identities and spread ideas whose consequences outpace their accuracy. In an age where language and images and claims are moving faster than ever, understanding how belief forms and how
it hardens is becoming something of a civic duty. If belief is the gateway between information and action, then learning how that gateway works is one of the most important challenges of our time. Will presumably never all agree on our beliefs, but a deeper understanding of this, I hope, can allow disagreement to remain tethered to reality and proportion
and humility go to eagleman dot com slash podcast. For more information and to find further reading, join the weekly discussions on my substack, and check out and subscribe to Inner Cosmos on YouTube for videos of each episode and to leave comments. Until next time, I'm David Eagleman and this is Inner Cosmos