🧠 The Uncomfortable Truth About What You Really Believe

Hummm, go deeper. Comic, haiku, mind maps, essay and more

Why your deepest convictions have more in common with falling in love than solving math problems

We tell ourselves a comforting lie about how our minds work. We like to imagine that our beliefs are the product of careful reasoning—that we weigh evidence, consider alternatives, and arrive at conclusions through some kind of internal cost-benefit analysis. It's a neat story that makes us feel rational, controlled, and fundamentally different from those "other people" who believe crazy things.

But here's the thing: that story is mostly bullshit.

If you've ever tried to change someone's mind about politics, watched a loved one stay in a toxic relationship, or caught yourself defending a position you know deep down is shaky, you've witnessed the gap between how we think belief works and how it actually works. And that gap isn't a bug in the system—it's a feature we've been systematically ignoring.

https://onlinelibrary.wiley.com/doi/10.1111/phis.12232

This is Heliox: Where Evidence Meets Empathy

Independent, moderated, timely, deep, gentle, clinical, global, and community conversations about things that matter.  Breathe Easy, we go deep and lightly surface the big ideas.

Thanks for listening today!

Four recurring narratives underlie every episode: boundary dissolution, adaptive complexity, embodied knowledge, and quantum-like uncertainty. These aren’t just philosophical musings but frameworks for understanding our modern world. 

We hope you continue exploring our other podcasts, responding to the content, and checking out our related articles on the Heliox Podcast on Substack. 

Transcript

0:25

Okay, let's unpack this. When you say you believe something, what does that actually mean? Most of us probably think, you know, it's pretty straightforward. Yeah, like you look at the facts, weigh the evidence. Right, and you decide something's true, and if new facts pop up, well, you change your mind. It's all very rational, logical. That's definitely the standard picture, certainly in a lot of philosophy. Belief is seen as this purely cognitive thing. defined by two key features. Which are? One, it's sensitive to evidence. You know, you update it when the facts change. Okay. And two, it reliably leads to action. Basically, what you believe guides what you do. Makes sense, kind of on the surface. But come on, think about real life, your own life. Have you ever believed something really stubbornly, even when evidence was like, piling up against it. Oh, definitely. Or said you believed one thing but then acted completely differently. Like, I believe I'm on a diet while, you know, eating a whole pizza. Exactly. And those kinds of cases, those real world examples, they show this gap, really. Yeah. A fascinating gap between that neat philosophical idea and... And the, well, the messy reality of how our minds actually work. And that gap is exactly what we're diving into today. We're doing a deep dive into a really thought-provoking paper by Miriam Schleifer McCormick. It's called Belief as Emotion. Right. And she looks right at that gap. And proposes something pretty radical, actually. What if we stopped thinking about belief as just this cold, logical calculation? And started thinking about it more like... Like a type of emotion. Whoa. Belief is an emotion. Okay. Yeah. Her argument is basically that the standard view, it's just too narrow. It ends up excluding loads of things we naturally intuitively call beliefs. Like the really deeply held ones or ones that just don't seem to budge when you show someone facts. or beliefs tied up with feelings. Exactly, though. So the mission here is to explore why that standard definition seems to fall short. And then to really get our heads around what it even means to call belief an emotion. It sounds counterintuitive at first. It does. But we'll explore the model she uses. And then we'll see how thinking this way might actually help us make better sense of those really tricky cases we all run into. Things like delusions or deep trust or even religious faith. Okay, so it's about building a bigger tent, maybe. A framework that can handle the full spectrum of how we hold things to be true, even the complicated bits. That's a good way to put it. It's about acknowledging the complexity. So where does the paper start? How does it challenge that standard view? Well, it kicks off by highlighting what it calls the bedfellow challenge. Beliefs bedfellows. Sounds intriguing. It refers to all those mental states that, you know, they look a lot like beliefs. We often call them beliefs in everyday life. But they don't play by those two standard rules. The evidence sensitivity and the action connection rule. Exactly. They're like beliefs, awkward cousins. The ones the standard definition kind of pretends aren't really part of the family because they don't behave properly. OK, so the standard view says if it doesn't change with evidence or doesn't reliably cause action, it's not really a belief. That's the gist of it. But the paper brings up some powerful examples that really make you think. question that neat dividing line. Meet us some examples. Who are these challenging bedfellows? Let's meet them. Okay, well take Anna. She's described as having cap grass syndrome. Right, the condition where people believe a loved one has been replaced by an imposter. Exactly. So Anna genuinely believes her husband isn't her husband, but an identical duplicate. Now think about the evidence. His appearance, his memories, things only her husband would know. It's overwhelming evidence that he is her husband. Right. It's massive counter evidence to her belief. Yet the belief persists. It's incredibly resistant to evidence. OK, so that clearly clashes with the evidence sensitive rule. What about the action part? Does she act like there's a dangerous imposter in the house? Well, that's the other tricky part. Not always in the way you expect if you truly unequivocally believe that. She might act differently towards him, yes, but maybe she doesn't flee. Or call the police constantly, or show the level of terror you'd predict. Huh. So there's a disconnect with the action-connected rule, too. It's not a straightforward link. Precisely. It complicates things. Wow, that's a really stark example. What's another one? Then there's Balthazar. He's standing on one of those glass bottom skywalks, way up high. Oh, I know the feeling just thinking about it. Right. So intellectually, Balthazar knows it's safe. Maybe he's read the engineering reports. He sees tons of other people walking on it. He has all the evidence for its safety. Logically, he believes it's safe. Yes, exactly. But what's happening? His body might be trembling, heart pounding. Maybe he's frozen, can't move forward. He feels intense fear. So his intellectual belief, this is safe, isn't translating into confident action or even feeling safe. Exactly. His state seems based on evidence, but his actions and feelings are totally out of sync with that supposed belief. The connection to action is unreliable there, too. Okay, so that's a clash between the intellectual side and the physical or emotional response. Or think about Charu. She trusts a lover who has cheated on her repeatedly. Yes, the triumph of hope over experience, perhaps. Perhaps. But despite this long track record, this evidence showing promises get broken, Charu finds herself believing they'll keep their word this time. That belief seems pretty resistant to the available evidence, wouldn't you say? Definitely evidence resistant. Okay, these are quite specific cases. What about more everyday, maybe less extreme examples? The paper also mentions David's religious belief. He believes in the God of the Bible. Now, he might be aware of historical findings or scientific claims that seem to contradict parts of the biblical narrative. Right. Potential counter evidence. Yes. And he might even explicitly say his belief isn't based on that kind of evidence anyway. It comes from faith or experience or something else. Which directly challenges that idea that belief must be purely grounded in factual, evidential reasons. It does. And the paper points out, look, these aren't just isolated, weird cases. Think about deep political convictions. Oh, boy. Often they seem driven more by loyalty, identity, values, strong feelings, rather than just a cool assessment of facts. And they can be incredibly resistant to counter arguments or evidence. Same with certain kinds of trust, maybe. Trusting a friend or a family member sometimes feels like it goes beyond just tallying up their past reliability scores. Absolutely. So all these cases, delusions, the skywalk fear, persistent trust... faith, political convictions they make, that neat, tidy standard definition of beliefs start to look, well, inadequate. Like it's missing a huge chunk of what it means to be human and hold something as true. Exactly. It feels like it's leaving too much out. So when philosophers are faced with these tricky bedfellows, what's the typical response? You mentioned the standard view wants to protect itself. Right. The common move, which the paper calls non-doxasticism. Non-doxasticism. Okay, breaking that down. Doxa is Greek for belief. So non-doxasticism means not beliefism. Pretty much. It basically says, okay, fine. These weird states, Anna's delusion, Palthazar's fear, Charu's trust, they don't fit my strict definition of belief. So they must be something else. Just invent a new category for them. Keep the belief category pure. Exactly. And philosophers have come up with various labels. Quasi-beliefs, aliefs, desires, in between states, by imagination, all sorts of things. The idea is to cordon off topics. true belief as this purely rational, evidence-tracking, action-guiding state. Yes, and often linked to the idea that proper belief should only respond to evidential, truth-seeking reasons. If feelings or values sneak in, it's disqualified from being a real belief. But the paper argues this whole strategy, this non-doxastic move, creates its own set of problems, Rick. Big problems. First, as you hinted at before, it risks massively depopulating the category of belief. Meaning, if we're that strict about the rules, we might end up saying that huge parts of what ordinary people think of as their beliefs, including irrational ones, biases, maybe even deeply held convictions, aren't actually beliefs, according to the theory. And why is that a problem? Why not just say, okay, philosophy uses belief in a very specific technical way? Because epistemology... The whole philosophical study of knowledge and rational belief needs to grapple with all the ways we hold things to be true Including the flawed or messy ways. I see if you define irrational beliefs out of existence You can't really study irrationality as a problem of belief anymore Exactly. You lose the tools to analyze why someone holds a delusion or how a political conviction might be epistemically problematic as a belief if you've already decided it's not a belief in the first place. It's like a doctor saying that's not a real disease because it doesn't fit my textbook definition of health. You need to categorize the problems to understand them. That's a great analogy. And a second issue is that this approach struggles to make important distinctions among these states it kicks out. Meaning it lumps them all together. Kind of. Is Anna's capgrass delusion the same kind of mental state as Charu's perhaps unwise trust in her lover or David's deeply felt religious faith? Intuitively, they seem different. Some feel pathological, some misguided, some maybe even justifiable in certain non-evidential ways. But if they're all just thrown into the not-belief bucket, it's hard to make those nuanced judgments about whether they're warranted or unwarranted, rational or irrational in different ways. Right. It flattens the landscape. And thirdly, as the paper notes, it just feels really counterintuitive when people themselves report having these beliefs. Like the person with capgrass saying, I believe my father is an imposter. Exactly. To then say, ah, but philosophically speaking, that's not really a belief feels like it dismisses their own description of their mental state. So trying to narrow the definition to keep it pure doesn't quite work. It creates as many problems as it solves. That's the paper's argument. People have tried tweaking the standard view, you know, saying belief just needs the capacity to revise or only needs minimal rationality. But the paper suggests these tweets still struggle with making those crucial normative distinctions, especially what one philosopher called the borderlands of belief. It just doesn't capture the complexity and at times messiness of our mental lives, as Eric Fidskabel puts it. OK, so if tightening the definition isn't the answer... What is? This is where the emotion idea comes in. This is the alternative proposal. Instead of shrinking the definition of belief to fit the theory, let's try broadening it. By thinking of belief as a type of emotion. Okay, I'm still trying to wrap my head around that. Beliefs feel like thoughts. Emotions feel like thoughts. I know it sounds like mixing categories, but it actually has some historical echoes. David Hume way back suggested belief is more properly an act of the sensitive than of the cogitative part of our natures. So feeling rather than just thinking. Kind of. And importantly, it connects to how many contemporary theorists understand emotions now. Which isn't just about raw feelings, right? You said they're more complex. Exactly. Most modern theories don't see emotions as just feelings. or just judgments or just motivations. They see them as integrated, blended states. Okay, unpack that. What does a blended state mean for an emotion like fear? Well, fear isn't simply the judgment that spider is dangerous. It's not just the feeling of your heart racing. It's not just the urge to run away. It's all those things, the representation. Spider is dangerous. The motivation flees, the physical and mental feeling, the butterflies, the dread, all bound together in one unified state. Fear. Got it. So emotions involve how we see the world, how we feel, and how we're moved to act all bundled up. Precisely. They have cognitive bits, feeling bits, motivational bits, all integrated. And the paper's core idea is that belief actually fits this blended model surprisingly well. Okay, how? Let's break down belief using that model. What are the components? All right, first, like emotions, beliefs are intentional. They are about something. Your belief is about the door being locked or about Paris being in France. Simple enough. Beliefs have content. Check. Second, beliefs are clearly linked to motivation and action. Maybe not always in that simple, predictable way the standard view demands, but they do shape how we navigate the world. Anna's belief about the imposter profoundly changes her interactions, even if not in the expected way. Fair point. Even messy beliefs influence behavior. Okay. Third, emotions usually have what philosophers call a formal object, the core thing they're about. Fears about danger, anger about offense or injustice, sadness about loss. Okay. So what's belief's formal object? What's it fundamentally about? Accuracy or truth. When you believe something, you represent that content as being accurate, as correctly mapping onto the world. That's the aim, the standard of correctness for belief. Right. Believing aims at getting things right. Makes sense. And fourth, and this is probably the most emotional part of the analogy the paper argues that belief is felt, representing something as accurate, taking it to be true, that feels a certain way. Okay, this is the bit I find trickiest. What does it feel like to believe, say, that Canberra is the capital of Australia? It doesn't feel like anger or joy. No, definitely not usually that intense. And the paper acknowledges this. It's often a much more subtle background feeling. Think about other low-level emotions like a quiet feeling of contentment when you're home or a faint background hum of anxiety about a deadline. You might not even notice them unless your attention is drawn to them. Or unless they change suddenly. Exactly. The feeling associated with belief might be like that. A subtle sense of correctness, certainty, settledness, or rightness that accompanies taking something as true. Hume used words like force, vivacity, firmness. Another philosopher, de Souza, talks about a feeling of rightness. A feeling of rightness. Hmm. So it's not necessarily a big flashing light, but more like a background hum that says, yep, this is accurate. That's a good way to think about it. It's part of the overall state of believing. And this feeling might vary in contensity, too. A weakly held belief might have a faint feeling, while a deep conviction might have a very strong feeling of rightness. Okay. Does this feeling have a positive or negative valence? Like, is it generally a good feeling or a bad feeling? The paper argues it has a generally positive valence. Not necessarily that the content of the belief is good news. Believing you failed a test feels bad because of the news. Right. But the state of believing itself, the sense of getting it right or representing accurately, has this intrinsically positive quality. Accuracy is seen as a good thing epistemically. So achieving it or feeling like you have has a positive feel or rightness to it. Interesting. So even if the news is bad, the feeling of knowing it, of it being settled and accurate, has a kind of positive epistemic stamp. Kind of. And crucially, this feeling isn't always a reliable guide to actual accuracy. That feeling of rightness can sometimes be misleading. Psychological research suggests it can link to things like overconfidence or resistance to rethinking something precisely because it feels so right. Ah, so the feeling can make us stick to beliefs even when maybe we shouldn't. It could be one factor, yes. So putting it all together, belief is seen as this blended state representing content, aiming for accuracy, having this feeling of rightness and being linked, though maybe complexly, to motivation and action. And viewing it this way as a blend allows for more flexibility. It doesn't demand that perfectly rational updating the standard view required. Exactly. It makes sense of why beliefs might sometimes be recalcitrant, why they might resist evidence just like emotions can. You might know the spider isn't dangerous, but you still feel scared and jump back. The feeling part of the state isn't aligning with the cognitive part. Okay, I think I'm getting the model. So how does this help us with those tricky bedfellows, delusions, trust, faith? Right, this is where the payoff is. By bringing these states into the category of belief as a type of emotion, we can stop arguing about whether they are beliefs and start analyzing them as beliefs but using this richer blended framework let's take delusions again anna believing her husband is an impostor or that really stark case the paper mentions r y who believed his father was an impostor and that he himself was dead How does the emotion view handle this? It would say these are beliefs. They definitely involve representing content. My husband is an imposter. Yeah. They're linked to action, even if the action seems strange or inconsistent sometimes. And crucially, they seem to involve an incredibly powerful, overwhelming feeling of truth or feeling of rightness about the content. But where does that feeling come from? Not evidence, obviously. Exactly. The paper suggests that in cases like delusions, the source of that intense feeling is pathological. It's arising from some dysfunction in the brain or mind, not from a reasonable assessment of evidence or even from understandable non-evidential reasons like love or hope. Ah, so the structure is like a belief representation plus feeling plus motivation. But the origin of the feeling component is the keto. difference. That seems to be the idea. We can identify them as beliefs, respecting the person's report, but we can normatively distinguish them, say they're irrational or unwarranted, precisely because of the pathological source of that dominant feeling component. It's like the feeling part has gone haywire and is overriding everything else. In the clinical context, the suffering it causes, the life disruption, also plays into why we see it as pathological, presumably. Absolutely. That broader context matters for our assessment. It respects the person's experience while still allowing for a critical evaluation based on the source and consequences. Okay, that makes sense for delusions. What about trust and faith? Like Charu trusting the unfaithful lover or David's religious faith? These can also be understood as beliefs within this blended framework, and it helps explain why the standard view found them so awkward. Because they often rely on non-evidential reasons. Exactly. Sherwood might trust her lover, not based on past evidence, which points the other way, but maybe out of love or a commitment to the relationship's value or even a desire to inspire trustworthiness in them. David's faith might be grounded in community, personal experience, or finding meaning, rather than purely historical or scientific evidence. Reasons the standard view tends to say shouldn't count for real belief. Right. But the belief as emotion view can accommodate these non-evidential reasons as potentially feeding into the overall state, particularly the feeling motivational aspects. And it suggests that relying on such reasons isn't automatically irrational. How so? When can non-evidential reasons be rational grounds for belief on this view? Well, the suggestion is it depends heavily on context and crucially on whether there's strong counter-evidence. trusting a friend based on your relationship history might be perfectly reasonable if there is no specific reason to doubt them trusting charru's lover given the history of deception seems much less reasonable because the counter-evidence is so strong so non-evidential reasons might be okay grounds for belief unless they require ignoring or flying in the face of significant evidence to the contrary. That seems to be the direction. It allows for a more nuanced assessment, we can ask. What kind of reasons are supporting this belief state? Are they evidential? Are they non-evidential, but perhaps warranted in this context? Yeah. Like trust in a reliable person. Or are they pathological, like in delusion? Yeah. Or are they perhaps non-evidential reasons that are being maintained despite overwhelming evidence, making the belief seem more like wishful thinking or self-deception? Okay, so it gives us more dimensions for evaluation than just does it track evidence? Yes, no. It's more like assessing an emotion. Is it fitting to the situation? Exactly the analogy the paper draws on. Think about anger. You're angry because you believe your neighbor dented your boat. Okay. We can assess that anger based on evidence. Did they actually dent it if not the anger is misplaced, like a belief based on false evidence? But we also assess anger based on context and proportionality. Even if they did dent it slightly, was flying into a rage appropriate. Are you someone who gets angry too easily? A dispositional issue. Is the anger damaging your life? I see the parallel. So for a belief like trust or faith, we can ask, is there strong evidence against it? If yes, holding it looks problematic. But even if evidence is neutral or absent, we might ask, are the non-evidential reasons like relationship value, hope, reasonable in this context? Is this belief serving a positive function or is it becoming harmful or detached from reality? Precisely. It lets us make distinctions. Believing in a vague spiritual force when evidence is inconclusive seems very different from believing, say, specific historical claims contradicted by strong evidence. The latter might require actively ignoring or suppressing evidence, making it lean closer towards the structure of delusion, even if the source isn't necessarily pathological in the clinical sense. So the huge benefit of this belief as emotion view is that it lets us keep all these messy states, delusions, trust, faith, convictions within the category of belief. It respects our intuition and how we talk. Yes. But it still gives us the tools to evaluate them, to critique them, to understand why some seem rational or warranted and others seem irrational, pathological or misguided. Exactly. It provides a framework for normative assessment that considers the whole blended state, the content, the feeling, the motivation, the reasons, evidential, non-evidential, pathological, the context. It embraces that complexity and times messiness we talked about earlier. Okay, so let's try and pull this all together then. The standard philosophical picture of belief, purely rational, evidence-driven, action-guiding... It's neat, but it struggles. It really does. It struggles with all those real world cases, the bedfellows, like delusions, stubborn trust, faith, strong convictions, where beliefs don't neatly follow the rules. And just kicking those states out, calling them not beliefs, doesn't really solve the problem. It feels unsatisfying and limits our ability to analyze them properly. Right. Sort of avoids the challenge rather than meeting it. So the paper's big idea. Think of belief as a type of emotion, not just a cold thought, but a blended state. A blend of representing something as accurate, having a kind of feeling of rightness or certainty about it, even if subtle, and being linked to motivation and action. And this richer picture, this belief as emotion model, it actually provides a framework that can include those difficult cases. Yes, it brings them back into the fold of belief while also giving us a more nuanced way to understand why they might resist evidence or connect strangely to action. And importantly, how we can still evaluate their rationality or appropriateness. It lets us consider the whole messy picture, the evidence, yes, but also the feelings, the non-evidential reasons, the context, the source of the conviction. Exactly. It allows for a much richer and arguably more realistic epistemology. Yes. And thinking about belief this way, it might just change how you listening think about your own beliefs, right? Are there things you believe, maybe not just because of a logical deduction, but because they feel fundamentally right? Or because they're tangled up with your identity, your relationships, your hopes? It invites that kind of self-reflection. And here's maybe a final thought to chew on. If believing is in some ways like having an emotion. Yeah. Does that mean we can evaluate beliefs not just for their truth, but for their overall appropriateness or health in a given context? Much like we might say someone's anger is disproportionate or their fear is unwarranted. Can you be criticized not just for getting the facts wrong, but for the way you hold the belief or the non-evidential reasons behind it? And could working on changing a deeply ingrained belief sometimes feel less like crunching data and more like, well, more like working on managing an emotional response or shifting a deeply rooted disposition? Wow. OK, that definitely gives us something to think about. Belief as emotion. A fascinating lens.