In complex, modern societies, ignorance and misperceptions are unsurprising. The puzzle - the fact that demands a deep explanation - is that some people hold accurate beliefs. Some thoughts:
Lots of academia is just people doing things that are technically impressive because it's easier to evaluate whether work is technically impressive than whether it actually contributes to our understanding of the world.
I *hate* videos like this, designed to mock ordinary people & make them look stupid. The most obvious point is that the interviews are cherry picked from a much bigger sample but made to look as if they're representative of people’s responses. But setting that aside…(1/5)
New essay: I argue that misinformation is often better viewed as a symptom of deep societal problems rather than their cause. When that’s true, interventions like debunking and censorship are unlikely to help – and might make things worse. (1/15)
"Everything is political" is a classic deepity. On one reading, it's true but trivial. On another, it's interesting but obviously false. These different interpretations create a misleading sense of profundity. It's Deepak Chopra but for people with PhDs in the humanities.
There's some misremembering on this point, so: The lab leak hypothesis that SARS-CoV-2 accidentally escaped from a lab absolutely was widely labelled "misinformation" and was widely censored.
I wrote a short piece on why the recent panic surrounding misinformation is misguided and how the concept of rationalisation markets is superior for understanding many forms of epistemic dysfunction in political and cultural media (1/17).
New open access paper: "The Marketplace of Rationalizations". The basic idea: Motivated reasoning is subject to a rationalization constraint: people can only convince themselves of beliefs for which they can find appropriate rationalizations. (1/16)
In general, people are highly rational, intelligent, and knowledgeable *about those aspects of reality that matter to them*. Using trick questions about pointless matters of fact doesn’t illustrate their stupidity; it just shows the ignorance of people laughing at them. (5/5)
It's tough putting together a 10-week philosophy of science course. I wanted to give a balance of traditional philosophy of science with more contemporary issues and debates, and also connect it to my own research interests. Here are the topics I landed on:
Much of social psychology seems to be premised on the bizarre assumption that what people really care about is not real-world outcomes but the state of their own mind: self-esteem, a positive self-image, dissonance reduction, feelings of control, reducing uncertainty, etc.
I'm teaching my philosophy of science course again this term 👇. Some brief thoughts on why Stephen Hawking was wrong about philosophy, five book recommendations for those interested in getting into philosophy of science, and the readings for my course:
First, being informed about pointless abstract facts about the world is a status symbol among highly-educated exam takers that constitute society’s elite. Most people don’t care, and have way more important things going on in their lives to bother acquiring this knowledge. (2/5)
Second, if you ask someone “Where is the Queen of England from?”, they will *reasonably* assume the interviewer is not mean-spirited & trying to trick them & make them look bad. So they probably assume (*reasonably*) that the answer can’t simply be in the question itself. (4/5)
“Continental philosophy” often strikes me as a religious project based on the hermeneutic parsing of sacred texts and a guru-based epistemology in which knowledge is advanced not by distributed, institutionally-scaffolded cognition but by decoding the profundities of sages.
Conservatives are incapable of producing good art because they lack empathy, which means they are incapable of creating anything which communicates to a larger audience. When they try to produce art, they only end up producing clumsy in-group signaling.
Rosa Luxemburg's insightful observation that "those who do not move do not notice their chains" applies to belief systems as well: it's only when your thoughts diverge from a dominant ideology that you notice the extreme social pressures with which it's held in place.
New essay: To avoid the charge that the post-2016 alarmism about misinformation has been a moral panic, researchers increasingly define misinformation to include content that is true but nevertheless misleading. I argue against this move: 1/16
What explains popular delusions and the madness of crowds? In this new essay I explore one important answer: Beliefs often function as *social signals* - and people care more about projecting a positive social image than about truth. Thread: (1/13)
This week, a group of prominent misinformation researchers published a commentary in Nature 👇on misinformation and misinformation research. I disagree with their analysis and explain why here:
Is misinformation a dangerous virus? Are we living through an infodemic? Is there a vaccine for misinformation? In this review of Sander van der Linden's (
@Sander_vdLinden
) new book 'Foolproof', I argue that the answer to these questions is "no". A thread:
Cool result. It seems like lots of research in psychology is converging on a simple but important finding: people are generally persuaded by rational arguments. I've updated my own views on this matter over the past year or two. My previous work was too pessimistic.
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
Most psychological work on human irrationality and self-deception is based on the "you can't handle the truth" model, which is greatly inferior to the "truth is often disadvantageous in social games involving persuasion, reputation management, and social signalling" model.
If you took those mocking such people and made them live their lives, they would crash and burn quickly because of how much local knowledge & understanding they lack. The mockers would seem like absolute idiots in spite of their high scores on “General Knowledge” tests. (3/5)
The UK has experienced its worst riots in over a decade. In a new article👇, I analyse the role of online misinformation. While many blame social media and call for increased regulation and censorship, this narrative is simplistic. I make five points: 1/11
Brilliant essay on misinformation by
@mnvrsngh
.
- Humans are not gullible
- Beliefs are extremely complicated. People often sincerely and passionately endorse highly compartmentalised beliefs.
- Misinformation is often a symptom of deeper social problems.
New piece: *Misinformation is not a virus, and you cannot be vaccinated againt it*. I explain why it's misleading to conceptualise misinformation as a contagious virus, and why there isn't good evidence that can people can be "inoculated" against it. (1/5)
My 'Socially Adaptive Belief' is now online. It explores the way in which bias, irrationality, ignorance, stupidity, etc., are often winning strategies in the strange social environments we confront as intelligent status-seeking coalitional primates.
No single article or book is ever that great. You learn by reading lots of books from diverse perspectives and building on knowledge incrementally over time. Assigning life-changing effects to single books often reflects a religious impulse at odds with a sensible epistemology.
There's an unfortunate tendency to think the truth is so self-evident that anyone who appears to disagree with you must really agree with you deep down and therefore be lying. Lots of talk about "disinformation" emerges from this tendency.
New essay 👇 exploring the evolutionary roots of human kindness and why the subtle incentives of reputation management explain why human altruism is both sincere and strategic. Our moral psychology is more Machiavellian than we like to admit. 🧵
Why do groups often embrace beliefs that strike outsiders as absurd? In a new paper (forthcoming in Mind and Language), I explore the increasingly influential idea that the absurdity of such beliefs makes them credible signals of ingroup commitment:
@ArthurCDent
It is strange, but LLMs pose a concrete challenge to his entire life's intellectual work (in linguistics and cognitive science, at least), so it's not that surprising, even if it is quite childish.
Martha Nussbaum and
@shadihamid
debate whether human overpopulation 👨👩👧👦or underpopulation 🚶♂️ will be the most important challenge of the future. Listen to the whole thing here 👉
"AI has taught us that human minds are hugely richer, and more subtle, than psychologists previously imagined. Indeed, that is *the* main lesson to be learned from AI." That seems right.
New preprint, 'Is the Brain an Organ for Prediction Error Minimizaton?' The conclusion: Maybe, but the free energy principle provides no reason for thinking that it is. More generally, it provides no support for any substantive theory of brain function.
New paper. I argue that there is no interpretation of the free energy principle on which it establishes both a first principles account of self-organisation and an interesting constraint on theorising in neuroscience.
Lots of scientific and popular articles on misinformation feature some variation on "misinformation is a growing problem" or "we are living through a misinformation age" and yet do not provide any evidence or even arguments in defence of these claims.
New post 👇. I argue: (1) The media rarely makes things up. (2) Media bias is nevertheless widespread. (3) Media bias is identified by biased individuals with beliefs acquired via biased media. (4) Misinformation is therefore hard to study objectively.
“Why is this even philosophy?!” asks person whose ideal of Real Philosophy is making narrow conceptual points within whatever extremely path-dependent conversation high-status members of the profession happen to be having in elite philosophy journals.
@dioscuri
I frequently encounter a mixture of incredulity/hostility for just calmly stating my view that markets - and more generally capitalism - are on net good.
Happy Darwin Day. Evolution by natural selection is the most important discovery in the history of our species - and yet most research in psychology and social science still proceeds as if the discovery had never been made. Sad.
There are many people who are more outraged by a well-meaning but conceptually confused Ted Talk than about the fact that Trump - someone Ramaswamy endorsed - literally tried to overturn an election result based on completely made-up, self-serving conspiracy theories.
NPR’s new CEO: “Our reverence for the truth might be a distraction getting in the way of finding common ground & getting things done.”
This gets to the heart of the cultural divide in the modern West: whether you believe truth is a priority or a hindrance
Human behaviour is profoundly shaped by status competition but people often strive to maintain a narrative in which what really drives them is not status but high ideals (Justice, Truth, Art, etc.) - precisely because such narratives reflect better on them and so promote status.
I've designed an undergraduate course, 'Science and Reason'. It will explore the scope and limits of science; how science relates to and differs from other modes of inquiry; and whether a scientific worldview threatens our commonsense understanding of reality and ourselves.
Self-deception is often a team project. It involves complex forms of social scaffolding and epistemic teamwork that generate collective delusions—bespoke realities—individuals could not achieve independently. New post 👇
Epistemic selfishness: biased reasoning that promotes one’s own interests but imposes costs on others. E.g., someone who adopts an inaccurate worldview for its emotional or reputational benefits, and then inflicts this worldview on others through activism.
“Analytic philosophy” often strikes me as a pseudo-scientific project based on imitating the superficial appearance of science (precision, clarity, mathematics) without including the thing that actually makes science powerful: its empirical engagement with the world.
The ideas you dislike - whether “wokeism”, conspiracy theories, misinformation, or religion - aren't “mind viruses” and don't spread via contagion. This framing rests on a stick-figure depiction of human psychology and communication that functions to demonise, not understand.
My PhD thesis, 'The Mind as a Predictive Modelling Engine: Generative Models, Structural Similarity, and Mental Representation', can be downloaded here:
I think publicly humiliating ordinary Trump voters by tricking them into revealing features of political psychology (partisanship, post hoc rationalisation, etc.) that are just as common among the liberal audience mocking & feeling superior to them is bad.
First lecture of the new academic year! An introduction to the history of AI, core concepts (computing machines, intelligence), the difference between good old-fashioned AI and artificial neural networks, Turing, and the Turing test.
The epistemic challenges of open societies, Part 2: Why the "marketplace of ideas" often functions like a marketplace of rationalisations designed to justify the favoured narratives of society's competing political and cultural tribes.
This article by Nikhil Krishnan, pointing out various fallacies and philosophical mistakes in constructionist theories of emotion, is *so good*.
via
@NewYorker
Much of the humanities is like this. "Everything is socially constructed", "The medium is the message", "Power is everywhere", etc. Good essay on deepities by
@DavidPinsof
here:
Can't stand stuff like this. Sanctimoniously mocking other people for the harmless status markers within their social milieu - driven by the absurd assumption that the mocker's own accent and self-presentation weren't shaped by exactly the same kinds of forces.
It's a good thing that no scientific field - such as, say, social psychology - has ever selected for sexy attention-grabbing results instead of accuracy.
Reminder than social media doesn't select for accuracy, but rather content that grabs your attention.
So be mindful that the content you see trending on here is often a poor reflection of reality.
A dilemma for misinformation research: define misinformation narrowly (e.g., as demonstrably false content) and it's rare; define it broadly (e.g., as misleading content) and it’s so pervasive that the concept is useless as a way of understanding information ecosystems.
In this new post I highlight my five favourite academic articles from last year, which range over topics like political ideology, religion, misinformation, reputation management, and intellectual humility: . Thread: 1/6
Walter Lippmann over one hundred years ago: There are no fingerprints of misinformation, and problems of belief formation in large-scale societies ultimately boil down to issues of trust and distrust. Succinct and correct 👍
I wrote a short piece on the social incentives that drive irrational beliefs. The basic idea: We are unconsciously biased towards beliefs that we benefit from propagating to others, and that signal our attractive traits and allegiances. (1/4)
via
@CAPDproj
I'm going to be writing here 👇 ≈weekly this year. In this first post I introduce the blog, explain its title 'Conspicuous Cognition', and outline why the pursuit of social approval drives the evolutionary weirdness of human behaviour and thought. 1/7
A lot of the energy and ingenuity left-wing intellectuals invest into explaining why people don't share their views ("ideology”, “manufactured consent”, “active ignorance”, "hegemony", etc) would be better spent simply trying to persuade people that those views are in fact true.
"Contrary to conventional wisdom, human brilliance emerges not from our innate brainpower or raw computational capacities, but from the sharing of information in communities and networks over generations." A false dichotomy.
People tend to be absurdly overconfident in their political beliefs.
This is not just misguided - it's harmful.
So what explains it? Here I explore some possible explanations:
I'm putting together a module for 3rd-year philosophy undergraduates on the general theme of "Politics, truth, and ideology". What would people include on the reading list? I'm interested in classic and modern works, as well as scientific in addition to philosophical ones.
A couple of years ago, I read Walter Lippmann's "Public Opinion" (1922) and it changed how I think about politics.
In this essay, I explore and evaluate Lippmann's radical critique of democracy.
Classifying Unherd as "disinformation" is wrong and indicative of how modern disinformation research often functions as a thinly veiled partisan project that simply reflects the personal views and preferences of researchers and organisations.
There should be stronger norms discouraging people from adopting 100% confidence in trendy activist stances on complex issues about which they are manifestly uninformed.
It's 2023 and people still talk as if they have privileged access to their motives. It would be better to admit that when we explain the reasons for our behaviour we are telling a story, and we typically have no greater access to whether that story is true than anyone else does.
I'm not much persuaded by Haidt's analysis, and I retweeted the Nature review. However, on reflection, the review is very uncharitable and unpersuasive. It doesn't engage with Haidt's arguments, and its identification of "real causes" is very implausible and overconfident.
A review in Nature, by
@candice_odgers
, asserts that I have mistaken correlation for causation and that “there is no evidence that using these platforms is rewiring children’s brains or driving an epidemic of mental illness.” Both of these assertions are untrue.
Have we evolved to love our ingroup? Did our ancestors go through an evolutionary process of self-domestication? Is the occasional genocide just an unfortunate by-product of how damn friendly we are? New essay 👇and thread: 1/15
If so, what might help? First, rather than investing so much into preventing the gullible masses from being brainwashed into holding bad ideas, it is far more important to win trust in institutions, including by *making them more trustworthy*. (11/15)
What are the best things you've read about the replication crisis? I'd be especially interested in something that would be appropriate for undergraduates to read as part of a course in the philosophy of science, but I'm interested in anything good on the topic.
Socialists like G.A. Cohen criticise capitalism for achieving social cooperation as a byproduct of self-interest. However, this is true of all complex cooperation. Capitalism is only unusual in making the role of self-interest undeniable. New post: 1/9
Interesting article. People don't reduce their demand for biased news even when they know it is biased - as long as the bias aligns with their own political views.
"The biggest red flag for me is when something evokes an emotion in you. That doesn’t necessarily mean that it’s misinformation, but it signals that somebody’s trying to manipulate your emotions." No it doesn't.
The central intuition driving the modern misinformation panic is that people – specifically *other* people 👇 – are gullible and hence easily brainwashed into holding false beliefs. This idea is wrong. (2/15)
When preferences for similar beliefs are widespread, this constraint gives rise to *rationalization markets* in which agents compete to produce justifications of widely desired beliefs in exchange for money and social rewards such as attention and status. (2/16)
New paper in which I try to answer the following question: How can you learn anything new from your imagination if you're the one deciding what to imagine?
Classic - and excellent - paper. "Beliefs are objects which provide values to their owners. The bases for these values have little to do with the probable truth of the beliefs. This is a crucial fact both psychologically and sociopolitically....
New draft: Scaffolding Motivated Cognition. The basic idea: In convincing ourselves of what we want to believe, we don't just rely on cognitive biases. As with most of cognition, we often make use of rich and complex forms of social scaffolding.
As with many culture wars topics, what masquerades as a controversy over a specific factual question - the origins of Covid - is really a dispute over the relative status and influence that should be awarded to different constituencies in society.
Scott Alexander is right about this 👇 It's an indictment of much of modern social science that this kind of view is now reflexively dismissed as a form of postmodernism when it's just what a scientifically informed approach to human psychology and epistemology implies.
The good thing about blogging is there's no editor to tell you to reduce the word count. You can write posts as long as you want and with the exception of losing 99% of your potential audience there's no downside.
People are systematically wrong about the world. This isn't primarily because they've been infected by “misinformation” but because of how human psychology interacts with the tendency of mainstream media to report a highly nonrandom sample of everything bad happening in the world
Last week the World Economic Forum published its 'Global Risk Report' identifying misinformation and disinformation as the *top global threats over the next two years*. In this essay I argue its ranking is either wrong or so confused it's not even wrong:
People are often motivated to attack, dominate, and eliminate others. To get away with this, they must recruit social support and maintain a reputation as moral and decent. Demonizing narratives often emerge to solve these problems. New essay👇