I’ve been on holiday, so this week I’m removing the paywall from an essay published earlier this year. New essays and other content will be back next week.
I’ve added “mostly” to the title to make clear that I’m not claiming there is literally nothing in the vicinity of wishful thinking. The argument is instead that wishful thinking is a much less important force in human psychology than many people—including many influential scientists and intellectuals—believe.
If you find the ideas in this essay interesting and want to explore them further, my new co-authored book, The Social Roots of Delusions, develops them in much greater depth (summary here), as do the articles listed under “Further Reading” at the end of the essay.
Thanks for subscribing!
Many people believe that we have a powerful tendency to convince ourselves of comforting falsehoods. We engage in wishful thinking, confusing our desires with our beliefs. We believe what we want to be true, not what is true.
More generally, we let our emotions distort our mental models of reality, embracing beliefs and belief systems that substitute reassuring myths for harsh realities.
Many also believe that this psychological bias is a significant force in human affairs. For example, it is supposed to explain why people fall prey to “positive illusions” (e.g., self-serving and self-aggrandising beliefs), why they convince themselves of religious fairy tales (the “opium of the masses”), and even why they accept absurd conspiracy theories, which allegedly reduce negative feelings associated with uncertainty and a lack of control.
This hypothesis—call it the “you can’t handle the truth!” model of human psychology—is so widespread that most people don’t even treat it as a hypothesis. It is viewed as a basic datum of the human condition, a powerful bias that might explain other things—self-deception, politics, religion, conspiracy theorising, and so on—but that couldn’t itself be seriously questioned.
For example, Scott Alexander simply defines motivated reasoning as “the tendency for people to believe comfortable lies, like ‘my wife isn’t cheating on me’ or ‘I’m totally right about politics, the only reason my program failed was that wreckers from the other party sabotaged it.’” In a post outlining his preferred explanation of this tendency, he notes that the “question – why does the brain so often confuse what is true vs what I want to be true? – has been bothering me for years.”
In contrast, I think Alexander has been bothered by a myth. There is no powerful tendency in human psychology to confuse what is true with what we want to be true. People do not generally convince themselves of comforting falsehoods.
Admittedly, there are some things in the vicinity of this tendency that are real. For example, we sometimes avoid acquiring or dwelling on information when we anticipate that doing so would be unpleasant, although this isn’t a very significant force in human affairs.
Moreover, I am not denying that motivated reasoning—the tendency for practical motivations and interests to distort our view of the world—is a powerful bias in human psychology. My claim is rather that the “you can’t handle the truth!” model completely misrepresents how motivated reasoning works in most cases.
Put simply: Although people often believe what they want to believe, they rarely believe what they want to be true.
Put another way: We often convince ourselves of falsehoods, but rarely reassuring or comforting falsehoods.
This is because motivated reasoning is driven by strategic, social goals rather than emotional ones. To understand how it works, you must replace the “you can’t handle the truth!” model with the “believing true things is often maladaptive in social games involving persuasion, reputation management, and status competition” model.
In this post, I will:
Describe the problems with the “you can’t handle the truth!” model.
Outline a rival social model.
Explain the former’s popularity.
As I will review, the social model is not original to me. It builds on the work of numerous scholars stretching back several decades. My goal is to draw these ideas together into a unifying framework and to highlight its theoretical and empirical support and explanatory power.
I will end by arguing that the “you can’t handle the truth!” model of human psychology is not just mistaken; it is pernicious. It encourages the view that when people accept “harsh” beliefs that they don’t want to be true, they are being rational and truth-seeking—even heroic. In reality, people are often motivated to convince themselves of negative, pessimistic beliefs, and it often takes courage and intellectual virtue to confront positive truths.
Against the “you can’t handle the truth!” model
Problem 1. Evolution
One classic problem with the assumption that humans have a general tendency to believe comforting falsehoods is that such a tendency would be extremely maladaptive. To make this concrete, consider other animals. Imagine if a chimpanzee didn’t bother competing for status because it convinced itself it was already the troop’s alpha, or if a wolf didn’t bother hunting for prey because it convinced itself it had a full stomach already.
These are extreme cases, but they illustrate a more general lesson: to navigate the environment, animals must be able to distinguish their beliefs and desires. This is equally true in human beings. To survive and thrive, we must be able to track the gap between what we want to be true and what is true so that we can take active measures to reduce it. This means that any systematic tendency to blur the distinction would have been ruthlessly selected against over the course of evolution.
Of course, humans sometimes have maladaptive tendencies, not least because evolution is not a perfect optimiser. So, these evolutionary reflections certainly don’t disprove the “you can’t handle the truth!” model. But they provide strong reasons for initial scepticism, especially if the model is supposed to identify a universal and powerful tendency in human psychology.
Problem 2. Implausibility
A more direct problem is that the model does a terrible job of explaining the most widespread and consequential belief systems where motivated reasoning seems to play a role.
Consider religion, for example. People often seem highly attached to religious beliefs and narratives, and motivated to hold on to them even in the face of strong contrary evidence. In other words, it is highly plausible that many people hold religious beliefs at least in part because they want to.
Nevertheless, it is difficult to see why people would want many popular religious beliefs to be true, many of which inspire feelings of terror, guilt, shame, anxiety, and dread. The pervasive belief in some version of hell (i.e., eternal damnation and torment) is the clearest illustration, but there are many others. Although people often talk about how religious ideas are “comforting” or “reassuring”, the literal content of many such beliefs is highly distressing.
Or consider politics. Once again, motivated reasoning seems to be on steroids in politics. This is not a domain where the average person seems to reason much like a disinterested truth seeker. And yet, once again, the idea that this is a domain where people convince themselves of comforting falsehoods is preposterous on its face. Look around! Do people really look like they’re prone to indulge in happy, positive worldviews? The reality is almost the opposite. Most people seem to be biased towards excessively negative, alarmist, and catastrophising belief systems. Many react with anger and hostility to the possibility that things are better than they thought—that trends are going in the right direction, that there has been progress—which is the opposite of what the “you can’t handle the truth!” model would predict.
Finally, consider conspiracy theories. Again, conspiracy theories often seem to be driven at least in part by motivated reasoning. Many people seem drawn to conspiracy theories. They seem dogmatically attached to them, even when they lack compelling evidence or clash with obvious facts. However, the case for the “you can’t handle the truth!” model of conspiracy theorising is even weaker than it is in the case of religion and politics. To put it mildly, it doesn’t seem very comforting or reassuring to believe that one’s country is run by a powerful, hidden cabal of evil elites conspiring against the public. The content of such a belief is extremely distressing! It is the kind of thing that wishful thinking would lead people to deny rather than affirm.
In response to such objections, proponents of the “you can’t handle the truth!” model often become creative. For example, many psychologists concede that conspiracy theories don’t seem to promote positive emotions on the face of it but argue that holding such beliefs nevertheless imposes a reassuring order and certainty on the world, increasing positive feelings of control and knowledge. In reality, the world is complex, confusing, and often random. Cartoonish narratives about conspiring villains make it more legible, replacing distressing uncertainty with comforting certainty.
This theory doesn’t make very much sense on its face. Do conspiracy theories about shadowy, powerful elite cabals really make the world feel more controllable and certain in ways that reduce anxiety? And if the goal is to produce reassuring feelings of certainty and control, surely there are countless other simplistic narratives that one could endorse that lack the distressing contents of most conspiracy theories? Moreover, why would humans have a general tendency to convince ourselves of falsehoods solely because they increase feelings of certainty to begin with? Once again, any such tendency would be extremely maladaptive. The goal of “reducing uncertainty” only makes sense if it leads to more accurate beliefs, not as an end in itself.
Another problem is that—unsurprisingly—the empirical evidence doesn’t support this picture at all. Not only do conspiracy theories not seem to have beneficial psychological consequences, but they “may even reinforce the negative experience of anxiety, uncertainty aversion, and existential threat”.
Some psychologists speculate that even though the function of conspiracy theories is to alleviate negative feelings of anxiety and uncertainty, they nevertheless fail to achieve this outcome. Just as you might eat a box of doughnuts to make you feel better, only for it to make you feel worse—now you’re sad and fat—people might convince themselves of conspiracy theories to make themselves feel better, even if in reality it has the opposite effect.
Of course, a simpler theory is that this is not what drives conspiracy theorising to begin with.
Problem 3. The Evidence
Relatedly, I am not aware of any compelling evidence that positive feelings or emotions associated with a belief are sufficient to drive motivated reasoning. For example, in a recent article titled “Good News Is Not a Sufficient Condition for Motivated Reasoning”, Michael Thaler reports results showing that although “most people expect good news to increase happiness”, this expected happiness does not elicit motivated reasoning. In other words, people are not biased to accept information merely because doing so would make them feel good.
When you reflect on it, this finding should not be surprising. Think about what you would believe if you were biased to accept beliefs merely because you want them to be true, and then compare those beliefs with what you actually believe. Now do the same for everyone else’s beliefs. You will find a huge disparity between the two.
Proponents of the “you can’t handle the truth!” model make a classic mistake. They observe certain beliefs (e.g., positive illusions or religious beliefs) and then develop a theory to explain what drives them. However, not only does this theory do a bad job explaining those beliefs, but its proponents rarely pause to reflect on what else we should observe if the theory is true, and whether those predictions align with what we observe in the real world.
A systematic tendency to convince ourselves of things that we want to be true predicts a very different world from the one that we actually observe. There are innumerable cases in which people could believe things that would produce positive emotions, but do not.
A Rival Social Model of Motivated Reasoning
These reflections suggest that despite the popularity of the “you can’t handle the truth!” model of motivated reasoning, it is not very plausible. Nevertheless, motivated reasoning itself seems to be a robust phenomenon. So, what is going on? Surely, you might think, if people are motivated to embrace specific beliefs, they must do so because they want those beliefs to be true?
No!
The most common mistake in theorising about motivated reasoning is to treat the following two claims as synonymous:
(1) People are motivated to hold specific beliefs.
(2) People are motivated to hold beliefs that they want to be true.
These are different claims. There is no logical inconsistency in the idea that someone might want to believe something that they do not want to be true, or even in the possibility of wanting to believe something that one wants to be false.
In fact, not only is this state of affairs coherent, but it is extremely common in the real world. To understand why, we need to abandon the “you can’t handle the truth!” model and focus on the strategic, social roots of motivated reasoning.
Socially Motivated Reasoning
When thinking about human psychology, there is a bizarre tendency, even among social psychologists, to overlook that we are extremely social animals. Most of our deepest, most powerful goals and motivations reflect the adaptive challenges posed by highly complex forms of social cooperation and competition.
This suggests that human sociality—the fact that we’re intensely social animals navigating intensely social environments—is an obvious place to search when trying to understand puzzling aspects of human psychology, including motivated reasoning. That is, instead of looking inwards to the desires and feelings of those who engage in motivated reasoning, we should look outwards to the incentives and constraints of the social environment they confront. As Philip Tetlock argues, we should consider “the social functions of thought and the embeddedness of human beings in relations with other people, institutions, and the broader political and cultural environment.”
When understanding motivated reasoning, two social functions are especially important:
Advocacy (persuasion, propaganda, deception)
Reputation management (signalling, prestige competition)
Advocacy
First, people are constantly looking for ways to promote their interests by spreading specific ideas and beliefs to others. In such cases, we behave like propagandists, advocates, or lawyers. Although this tendency can be conscious and self-aware, it is typically instinctive and connected to motivated reasoning. That is, our brains are constantly but unconsciously tracking which ideas would promote our interests, and we are disposed to sincerely internalise such ideas, to become convinced of the very beliefs we want to convince others of.
Robert Trivers was the first social scientist to make a rigorous case for this feature of human psychology, although the core phenomenon—our tendency to internalise beliefs that we’re motivated to spread or advocate for—was also recognised earlier by thinkers such as Nietzsche and Hannah Arendt.
Although much remains unclear about how precisely this psychological process works and how it evolved, it is a relatively well-established tendency that provides a powerful explanation of many examples of motivated reasoning.
Consider “demonising narratives”, for instance. Beliefs as diverse as witchcraft accusations, dehumanising myths about rival outgroups and minorities, and many conspiracy theories all share something important in common: they denigrate target groups in ways that justify expressions of hostility towards them, including, in extreme but relatively frequent examples of genocide throughout human history, their outright elimination. In other words, the propagandistic motives underlying such belief systems are relatively transparent.
As Hugo Mercier notes, this suggests that Voltaire’s famous quote, “Those who can make you believe absurdities can make you commit atrocities,” gets things backwards; it is the desire to commit atrocities that motivates people to embrace absurdities.
For a very different example, consider widespread beliefs in supernatural rewards (e.g., heaven) and punishments (e.g., hell) that constitute the core of most religious belief systems worldwide. As Léo Fitouchi, Manvir Singh, Jean-Baptiste André, and Nicolas Baumard document in an insightful recent article, these beliefs plausibly emerge and stabilise as a way of policing and manipulating other people’s behaviour.
If you want to encourage members of your community to be more cooperative and norm-following, you have a strong incentive to spread beliefs about supernatural punishment, especially if you inhabit a world lacking institutions and legal systems that enforce prosocial behaviour. Over time, these incentives, when coupled with people’s disposition to internalise beliefs they’re motivated to spread, give rise to religious belief systems that look designed to support systems of cooperation, which—probably not coincidentally—constitutes one of humanity’s central adaptive challenges.
Reputation Management
A distinct social motivation at play in motivated reasoning stems from various processes related to reputation management. In such cases, the believer’s goal is advertisement, not advocacy. The goal is to cultivate and advertise—that is, signal—traits and loyalties that contribute to a socially lucrative reputation.
Of course, this goal drives vast amounts of human behaviour, but research also suggests that it seeps into the beliefs and broader worldviews we develop. Put simply, we convince ourselves of beliefs that look good on us—that send the right messages about the kinds of people we are and the social allegiances we have. If there is a metaphor here, Robin Hanson’s suggestion that many beliefs function like “clothes” is helpful. Although some decisions about clothing are shaped by considerations of simple functionality (e.g. keeping oneself warm), most decisions are driven by considerations of reputation management. People want to show off, fit in, advertise their group memberships and allegiances (e.g., social class and political identities), and so on.
Many beliefs are similar.
For example, people plausibly embrace political, religious, and cultural narratives at least in part to signal their tribal loyalties. In some such cases, the contents of the beliefs themselves can be arbitrary. All that matters is that they are identity-defining, demarcating the in-group from outsiders. In other cases, affirming specific beliefs bears a more direct relationship to things the group is likely to care about, such as with narratives that signal group allegiance simply by valorising the in-group and denigrating its rivals.
There are also plausible examples of belief signalling focused not on loyalty to specific people or communities but on other traits and qualities. For example, just as individuals can be motivated to spread beliefs about supernatural punishment to persuade others to be more cooperative, they can also be motivated to embrace such beliefs to signal their own cooperative tendencies. In other words, I can benefit myself by persuading you that God punishes selfish behaviour, but I can also benefit myself by persuading you that I think God is always watching, because then you will be more likely to trust me.
Another possible example is how paranoid and conspiratorial narratives can signal social vigilance for exploitation and manipulation. In The Godfather, Don Corleone says to the heads of the other mafia families, “I’m a superstitious man, and if some unlucky accident should befall [my son], if he should get shot in the head by a police officer, or if he should hang himself in his jail cell, or if he’s struck by a bolt of lightning, then I’m going to blame some of the people in this room.”
By broadcasting such suspicious and superstitious tendencies, Corleone signals an important message about himself, an act that might tell us something about the signalling motivations fuelling paranoid, conspiratorial tendencies in the general population more broadly.
“Active Belief”
So far, I’ve focused on how motivations to propagate specific messages and signal inner qualities and loyalties can drive people to internalise specific beliefs. But as Will Storr argues in The Status Game, socially-motivated beliefs and belief systems rarely take the form of passive interpretations of reality. They often involve what Storr calls “active belief”, a cluster of familiar human tendencies that includes evangelism, proselytising, and hunting down heretics to be publicly punished. In such cases, Storr argues, active belief becomes a means of advancing the status and influence of belief-based communities and of winning status within them.
These tendencies can be linked to motivations to propagate beliefs and to embrace them for social signalling purposes. For example, when communities benefit from spreading specific belief systems or narratives, those who pay the costs of spreading and enforcing those beliefs—who exhibit active belief—are acting to benefit others, which is a sure route to winning prestige in human communities. Similarly, when beliefs become signals of group membership and allegiance, active belief constitutes a costly and thus credible signal of commitment to such beliefs and, therefore, to the broader community.
The Social Roots of Irrational Beliefs
Of course, these remarks barely scratch the surface of these complex issues. Nevertheless, several things are noteworthy about all of this.
First, this social model is completely different from the “you can’t handle the truth!” model. If this framework is correct, people’s motivations bias their beliefs not because of any desire to believe comforting falsehoods, but because strategic, social objectives connected to advocacy, signalling, and status competition conflict with the disinterested pursuit of truth. For this reason, the social model is perfectly consistent with cases in which people appear highly motivated to embrace beliefs that they don’t want to be true. People don’t want to believe that most demonising narratives are true—it’s distressing to believe that an outgroup is extremely dangerous and threatening, for example—but such beliefs can nevertheless advance people’s social interests.
Second, precisely because this social model is connected not to in-the-head emotion regulation but to concrete, social outcomes that carry huge practical importance for people’s survival and success, it has much greater evolutionary plausibility than the “you can’t handle the truth!” model. From an adaptive perspective, it never makes sense to replace one’s beliefs with one’s desires, but it might make sense to sacrifice some amount of accuracy in one’s model of the world in the service of winning social approval, trust, and prestige.
Third, although the empirical evidence concerning motivated reasoning is by no means dispositive, it seems to support this social model much better than the “you can’t handle the truth!” model. For example, the finding that good feelings alone don’t suffice to initiate motivated reasoning, but partisanship and self-image concerns do, is exactly what one would expect on the social model. Spreading self-serving and partisan narratives can advance one’s self-interest in ways that believing things merely because they are comforting and reassuring never will.
The False Allure of the “You Can’t Handle the Truth!” Model
If the social model is correct, why is the “you can’t handle the truth!” model so much more popular?
Robert Trivers gives a characteristically cynical answer. Whereas the social model suggests that humans are savvy operators attempting to manipulate others for social goals, the “you can’t handle the truth!” model suggests that motivated reasoning involves a set of defence mechanisms by which we protect ourselves from threatening truths.
He writes, “A defensive view of self-deception is congenial to an inflated moral self-image—I am not lying to myself the better to deceive you, but rather I lie to myself to defend against your attacks on myself and my happiness.”
So, one possibility is that people are drawn to the “you can’t handle the truth!” model precisely because it is more consistent with the self-image they want to project to others. The idea that we are constantly seeking to embrace or spread beliefs that help us win status and approval makes us seem vain and Machiavellian, so we are motivated to reject it.
Another analogy with clothing is illuminating here. A popular slogan in the world of fashion is, “Dress to express, not to impress.” In other words, the reason why people spend vast amounts of money, time, and energy on fashion is not out of any small-minded, competitive desire to win social approval. It is driven by far nobler purposes of “self-expression” and “authenticity”.
The fact that such a preposterous idea could win widespread acceptance suggests that Trivers is onto something. People go out of their way to deny that they are trying to impress others, even when the fact that they are trying extremely hard to impress others could not be any more obvious. Perhaps our “folk” models of motivated reasoning are biased in a similar direction.
Another possible (and complementary) explanation is that the “you can’t handle the truth!” model is simply more intuitive than the social model. It is relatively easy to understand how people’s emotions could bias their beliefs because emotions are, in some sense, in their heads where the beliefs are. In contrast, the social environment—that is, other people’s responses to what an individual says and believes—is outside the head. So, how could features of the social environment reach inside the brain to bias people’s beliefs?
Of course, it is easy to understand this if people are consciously thinking about the social environment, but the whole point about motivated reasoning is that it is not conscious in this way. People aren’t consciously reasoning, “If I believe X, this will have desirable social consequence Y.” But then, how is socially motivated reasoning supposed to work? The whole thing seems strange and counterintuitive.
The mistake that people are making here is analogous to the one people make when they assume that sex can’t have evolved reproductive functions because people aren’t typically consciously thinking about reproduction when they seek out and have sex. They are just thinking about how good sex feels. This is true, but the primary reason why sex feels good—and more generally the reason why people’s sexual psychology works the way it does—is because sex historically led to reproduction, so natural selection gave rise to psychological mechanisms that motivate it.
Likewise, the social model of motivated reasoning needn’t posit that motivated reasoning involves complex, conscious calculations of the social costs and benefits of different beliefs. Instead, the idea is that human psychology evolved—and, over the course of an individual lifetime, develops—in such a way that people are instinctively drawn to beliefs that advance social interests connected to persuasion and reputation management.
This clarification is important because it shows that emotions and feelings can still play a role in the social model of motivated reasoning. For example, it likely feels good to affirm beliefs that signal membership of and allegiance to one’s community, or to spread beliefs that demonise one’s rivals and enemies, and these feelings likely play a role in biasing how people seek out and process information in such cases. In this sense, feelings remain part of the proximate causal mechanism underlying motivated reasoning. However, if the social model is correct, these feelings don’t result from aligning beliefs more closely with what the individual wants to be true; they result from aligning beliefs more closely with strategic, social objectives that we evolved to pursue.
Finally, another (again, potentially complementary) explanation for why the “you can’t handle the truth!” model is so popular is that, in some cases, people really are motivated to believe things that they want to be true. For example, people are often motivated to embrace “positive illusions”, which involve self-serving and self-aggrandising beliefs that most people want to be true of themselves (e.g., that they’re virtuous and handsome). Similarly, many religious narratives embody intuitions about divine justice—namely, that people who do commit terrible acts will ultimately be punished, even if they escape human accountability—which many people find comforting. The “you can’t handle the truth!” model seems to fit these examples, so it’s understandable that some people take it seriously.
However, it’s important to note that the social model of motivated reasoning also fits these cases. The social model doesn’t deny that sometimes people convince themselves of things that they want to be true. It denies that this is the reason why they convince themselves of such things. Sometimes, socially adaptive beliefs happen to coincide with comforting beliefs. When that happens, there will be an overlap in the predictions of the two models. For example, flattering self-narratives are predicted by both the hypothesis that people believe what they want to be true and the hypothesis that people internalise beliefs they are motivated to spread to others.
This means that if people restrict their attention to only these cases of motivated reasoning, the flaws of the “you can’t handle the truth!” model won’t be apparent. To appreciate these flaws and the superiority of the social model, you must focus on all the cases in which people are also motivated to embrace beliefs that they don’t want to be true, as well as on all the beliefs people would hold if we were disposed to convince ourselves of reassuring falsehoods.
In other words, you need to engage in some sophisticated forms of scientific reasoning, which have been absent in most research on motivated reasoning.
Final Thoughts
Suppose that you’re convinced by the arguments in this essay. Nevertheless, you might reasonably ask, “So what? Why does any of this matter?”
The simplest and most obvious answer is that motivated reasoning is an extremely consequential feature of human psychology and societies. As Adam Smith said, self-deceit is “the fatal weakness of mankind” and “the source of half the disorders of human life.” That was an exaggeration, perhaps, but an understandable one. This makes the task of developing an accurate understanding of motivated reasoning of great importance, both in the psychological and social sciences and in matters of practical concern. For example, if we want to reduce the impact of motivated reasoning on what people believe in domains like politics, we must understand its causes.
Relatedly, I suspect that the popularity of the “you can’t handle the truth!” model does a lot of damage in political discourse. If your model of human psychology is that people routinely convince themselves of reassuring falsehoods, you will suspect motivated reasoning whenever people hold “positive” or “optimistic” beliefs about the world or about progress, and you will be inclined to treat those who hold bleak and pessimistic beliefs as courageous truth-seekers—as bold thinkers willing to overcome self-deception and confront harsh realities.
These reactions are misguided. There is no general bias in human psychology toward being convinced by reassuring falsehoods. In contrast, domains like politics often generate powerful social incentives to spread and embrace negative, alarmist narratives. For example, such narratives can function to demonise rivals and outgroups, exaggerate threats and dangers that people claim to fight against, and signal intellectual seriousness and public concern.
Against these tendencies, it often takes courage and virtue to point out progress and positive trends, to push back against alarmism, and to argue that the world and the outgroup may not be as threatening as people claim.
In our ultra-social species, the fight against motivated reasoning isn’t a fight against comforting falsehoods; it’s a fight against the pressures and incentives of social life.
Further Reading
Robert Trivers passed away shortly before this essay was published. Nobody else has been more influential in the development of the social model of motivated reasoning outlined in this article and in my ideas about human psychology more broadly than he has. His book, The Folly of Fools, is a good place to start to learn more.
Beyond that, there are far too many thinkers, articles, and books to cite as influences on the social model of motivated reasoning, but a helpful initial list to start digging deeper includes the following:
Philip Tetlock, “Social Functionalist Frameworks for Judgment and Choice”
Robert Kurzban, Why Everyone (Else) Is a Hypocrite
Kevin Simler & Robin Hanson, The Elephant in the Brain
Will Storr, The Status Game
Nichola Raihani, The Social Instinct
Hugo Mercier, Not Born Yesterday
Eric Funkhouser, “Beliefs as Signals: A New Function for Belief”
Léo Fitouchi, Manvir Singh, Jean-Baptiste André & Nicolas Baumard, “Prosocial Religions as Folk-Technologies of Mutual Policing”
David Pinsof et al, Strange Bedfellows: The Alliance Theory of Political Belief Systems
I’ve tried to develop these ideas in a series of academic articles:
EDIT: Economic psychologist Joël van der Weele contacted me to argue that my reading of the empirical literature doesn’t do justice to the evidence for hedonically-motivated cognition, and asked that I add this review article to the reading list for those interested in evidence against my claims: The renaissance of belief-based utility in economics