My five favourite academic articles from 2023

Evolution, Ideology, Religion, Misinformation, Reputation, Humility

2023 was a great year for academic research. In this post, I’ll list five of my favourite articles.

First, some context: As an academic, I read lots of academic articles every year. Although I’m a philosopher, the research I do is anti-disciplinary and naturalistic, and I have wide-ranging interests, so most of the articles I read are in psychology and social science.

The overarching focus of my research is on why human beings believe what they do, especially when they seem to believe things that are irrational or weird. So I try to keep up-to-date with research on topics surrounding ideology, delusions, religion, self-deception, irrationality, propaganda, and misinformation.

Here are some of the best things I read on these topics in 2023.

1. Ideology

What gives rise to political ideologies? What explains the specific contents they have? Why do conservatives endorse a different ideology to liberals?

Many influential answers to such questions attempt to trace ideologies to abstract values (like equality, freedom, and tolerance). According to David Pinsof, David Sears, and Martie Haselton, that approach is misguided. In ‘Strange Bedfellows: The Alliance Theory of Political Belief Systems’ (non-paywalled version here), they offer a more cynical theory.

Political belief systems, they argue, emerge from two basic foundations.

First, there are alliances, people and groups who team up in specific times and places to promote their collective interests through the political process.

Second, there are propagandistic tactics with which partisans frame and interpret reality in ways that promote their alliance’s interests and rally opposition to their rivals.

Given this, the main difference between, say, conservatives and liberals doesn’t involve differences in fundamental values. It’s much simpler: whom they treat as their allies and rivals in competition for power, status, and other goods.

The result is political belief systems specialised not for Truth or Justice but for promoting and justifying the interests of some alliances over others. As Pinsof and colleagues detail, this explains why political ideologies frequently look less like coherent philosophies than bundles of rationalizations, embellishments, moralizations, ad hoc justifications, and gross hypocrisy.

The theory is a great achievement. It’s impressive to come up with a genuinely novel theory of something - political belief systems - that’s received so much attention. It’s even more impressive to produce a theory that’s evolutionarily and psychologically plausible, parsimonious, and capable of explaining a wide range of findings.

Admittedly, one reason I like the theory so much is because it aligns with my own perspective on political psychology. In fact, Pinsof emailed me the article when I was finishing the revisions on a paper exploring how political cognition is shaped by motivations to promote and justify coalitional interests. Although the articles have a different focus, they’re complementary.

• Note that ‘Strange Bedfellows’ was accompanied by numerous commentaries (to which I contributed) and a response by Pinsof and colleagues.

2. Religion

Throughout human history and across incredibly diverse cultural contexts, people endorse religions and proto-religions organised around beliefs in supernatural agents and forces. Why?

Under the assumption that such agents and forces don’t exist - given that religions typically contradict each other, most can’t exist - this human tendency is puzzling. Why do humans spend so much time and energy engaging with mysterious, non-existent entities and cosmic fairytales?

Historically, many social scientists and philosophers (Durkheim, Marx, Freud, etc.) grappled with this puzzle, and it remains an enormous area of live research. In my view, the most persuasive theory comes from an article by Léo Fitouchi, Manvir Singh, Jean Baptiste André, and Nicolas Baumard.

The article is ‘Prosocial religions as folk-technologies of mutual policing’. (Note that it was uploaded as a draft last year; it’s not actually published yet). It develops a classic perspective shared by figures like Durkheim: that much of religion is connected to problems of cooperation, one of the central challenges of human social life.

The motivation for this perspective is obvious. Many aspects of religion, including the postulation of moral rules supported by supernatural surveillance (e.g., “God is always watching”) and enforcement (e.g., heaven, hell, karma, etc.), appear to be designed to promote cooperation.

Given this, many theories assume that prosocial religions emerge because they perform this function well. For example, one influential theory holds that prosocial religions evolve and proliferate because cultural groups that happen to stumble on such religions outcompete cultural groups that stumble on non-social or antisocial religions.

This kind of theory confronts at least two basic problems, however. First, even though religions often appear to be designed to promote cooperation, it’s not actually clear that they in fact make people more cooperative.

Second, the mechanisms posited by such theories are often pretty opaque. For example, the idea that prosocial religions spread because groups embracing prosocial religions outcompete other groups offers no explanation of why prosocial religions emerge to begin with.

In response to these and other problems, Fitouchi and colleagues make a very simple but ingenious move. They propose that prosocial religions emerge because individuals (a) believe such religions will promote cooperaton and (b) are motivated to make other people more cooperative.

In other words, religions evolve as people strive to manipulate other people into being more cooperative. At the same time, because people want to be seen as cooperative, they’re also motivated to endorse religious beliefs to signal that they’re cooperative.

This theory has two important advantages over previous theories that link religion to cooperation.

First, the mechanism it posits is extremely plausible and unmysterious. We know that people are motivated both to make other people more cooperative, and to appear cooperative to others. We can therefore explain what’s going on here in terms of the motives and interests of strategic individuals. We don’t have to invoke strange, underspecified forces like “cultural group selection”.

Second, this theory doesn’t require that religions in fact promote cooperation, which is controversial. All it requires is that people believe prosocial religions promote cooperation, and that’s uncontroversial. (In many places throughout the world, for example, atheists are one of the least-trusted groups).

Obviously this very brief summary doesn’t do justice to the nuances and complexity of the article, which should be essential reading for anyone interested in psychology and social science.

• See also ‘Supernatural punishment beliefs as cognitively compelling tools of social control’ by Léo Fitouchi and Manvir Singh.

3. Can people be “inoculated” against misinformation?

Since 2016, Western democracies have been gripped by a widespread panic about misinformation. Although it’s true that misleading content in politics and media is widespread and harmful - it always has been - I’m a critic of the ways in which the liberal establishment has framed misinformation as a convenient catch-all explanation of various worrying political trends (see here, here, here, and here).

One very popular idea in misinformation research is that people can be inoculated against misinformation by playing games or watching videos that teach them about common “manipulation techniques”. These includes things like the use of emotional language, polarising rhetoric, and logical fallacies. According to advocates of this approach, its success is supported by numerous experimental studies.

In ‘Gamified Inoculation Interventions Do Not Improve Discrimination Between True and Fake News’ (non-paywalled version here), Ariana Modirrousta-Galian and Philip Anthony Higham challenge this view. They reanalyse results from prior studies and uncover something interesting: such inoculations do not help people to discriminate between true and fake news items; instead, they simply make participants more likely to classify all items (whether true or false) as false.

This matters. Much of modern misinformation research is founded on the idea that human beings are gullible - that they’re too ready to accept false information. In fact, the problem seems to be the exact opposite of this. In general, human beings are sceptical social learners, if anything too pig-headed - placing too much weight on their own beliefs relative to those of others - than too credulous. Moreover, most of the news that people encounter is not fake news.

Given this, it’s a really big problem if one of the most influential interventions in misinformation research makes people more likely to rate news in general as false or deceptive.

By drawing attention to this issue, Modirrousta-Galian and Higham’s article is a very important contribution (and one that I draw on in my own, broader critique of inoculation theory here and here).

4. Misinformation on misinformation

Another great article on misinformation is ‘Misinformation on misinformation’ by Sacha Altay, Manon Berriche, and Alberto Acerbi. I’ve drawn on ideas in this article and by other works of the authors (e.g., here, here, and here) many times in my own essays on misinformation.

They argue that much of the hype and hysteria surrounding misinformation rests on the following six misconceptions:

1. Misinformation is just a social media problem.

2. The internet is rife with misinformation.

3. Falsehoods spread faster than the truth

4. People believe everything they see on the internet

5. A large number of people are misinformed

6. Misinformation has a strong influence on people’s behavior.

I disagree with a few things in the article. For example, generalisations about misinformation (e.g., its prevalence and effects) depend on how the concept is defined. Although they acknowledge this, I think some of the things they say about the relative rarity and symptomatic character of misinformation rests on a narrow definition of misinformation that misses many of the ways in which communication can be misleading and harmful.

The same is true of the question concerning whether large numbers of people are “misinformed.” That depends on what it means to be misinformed. If you focus on narrow factual matters, then it’s true that people tend to be ignorant rather than actively mistaken about the world. But in terms of the more general ways in which people understand social, political, and economic reality, I think current research tends to greatly under-estimate how distorted and inaccurate most people’s mental models of the world are. (I will be writing about this in the future).

Nevertheless, it’s a great paper - one of the best articles written in this general area - and I recommend it to anyone interested in misinformation or popular narratives about misinformation.

5. Reputation and Rationality

There’s a vast body of research on intellectual humility, the willingness and ability to acknowledge one’s fallibility and ignorance. For example, there’s research in philosophy that views intellectual humility as an epistemic virtue that promotes people’s ability to acquire knowledge, and there’s research in psychology that explores individual differences in intellectual humility and the correlates of such differences.

In ‘The Reputational Benefits of Intellectual Humility’, Mia Karabegovic and Hugo Mercier advance a different perspective. They explore how appearing - or failing to appear - intellectually humble has important reputational consequences.

Given this, people can have reputational incentives to be intellectually humble - and to be intellectually arrogant. For example, intellectual humility can signal that one is an attractive and competent source of information. At the same time, a conspicuous lack of intellectual humility can be used to signal dominance.

This general perspective seems really important.

First, you can’t understand human psychology and social life without understanding how much energy people invest in reputation management.

Second, there’s good evidence that people’s intellectual traits - their intellectual virtues and vices - play a big role in their ability to view the world accurately and make a positive contribution to public discourse.

For example, research by Philip Tetlock on the characteristics of reliable forecasters shows that good thinking is rooted in intellectual humility, curiosity, a willingness to acknowledge uncertainty, the motivation to integrate diverse perspectives rather than blindly follow ingroup opinion, and - most importantly of all - the rare motivation to treat beliefs as “hypotheses to be tested, not treasures to be guarded”.

Given this, exploring how reputation management shapes the kinds of intellectual virtues people cultivate - and fail to cultivate - seems like an extremely important and under-researched project.

• See also ‘Reputation Management and Cultural Evolution’ by Hugo Mercier.